COVID-19 and acute kidney injury in German hospitals 2020

22

23 The SAR S-CoV-2 pandemic is a majo r challenge for patien t s , healthcare p rofess ion a ls , and 24 populat i o ns worldwide. Whil e ini t i al repo rt ing focused mainly on lung involv ement, th e 25 ongoing pandemic sho wed t hat m ultiple or gans ca n be inv o l ved, and p r ognosi s i s la r gel y 26 influenced by m ulti-organ involvem ent. Our aim w a s to obtain nationwide retr os p ec t iv e 27 populat i o n -bas ed d a t a on hospitaliz ation s with COVID -19 and AKI . 28

& Methods 29 We perf ormed a quer y of G -D RG data for the y ear 2020 v ia the Instit u t e f or the ho s pit a l 30 rem unerat ion sys t em (Institut fü r d as Ent gel t system im Krankenhaus Gmb H, InEK) data 31 por tal and th erefo r e inc lud ed hospitaliz ation s with a s econdar y dia gno si s of RT-P CR proven 32 CO VID -19 inf ec t ion, a ged over 15 yea r s . We included hospitaliz at ions with acute ki d ney 33 injury (AKI) s t ages 1 t o 3. A ge-s pe cific and a ge-stand a r diz ed ho spitaliz ation and in-ho s pital 34 mor tali ty r ates (A SR) per 100.000 per s o n year s w er e calcula t ed, wit h t he German 35 populat i o n o f 2011 as t he s t a n dard. 36

37 In 2020, there wer e 16.776 .845 ho sp italizations in G er man hospitals. W e detected 154.170 38 hospitaliz at ions wit h RT-PCR proven COVID -19 diagno si s . The age-standar diz ed 39 hospitaliz at ion rat e for COVID- 19 in Germany wa s 2 32,8 per 100. 000 per son years (95% CI 40 231,6-233 ,9) . The highe st propor tion o f hospitalizations a ssociated with CO VID- 19 were in 41 the age group over 80 years. AK I w as diagno s ed in 16.773 (1 0,9%) o f th e hospitaliz at ions 42 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 3 with C OV ID-19. The relative r is k o f AKI for males was 1,4 9 (95 %CI 1, 44- 1,53) compar ed to 43 females. Renal r eplac em ent t herapy ( RRT) was per formed in 3.443 hospit aliz ations, 20 ,5 % 44 of t he hos p i t ali zations with A KI. For all hos pit ali zations with C OVID - 19, the in-h os pit a l 45 mor tali ty amount ed to 19,7% (n= 30 . 300) . The r elative risk f or in-hospital mor tali ty wa s 3 ,87 46 (95%CI 3 ,80-3 ,9 4) when AKI oc c u rred. The age-standar diz ed hospitalization rat e s for CO V ID-47 19 took a bimodal cour s e during the obse r vation period. The fir s t pea k oc curr ed in Apr i l 48 (ASR 23,95 per 100.0 00 per son year s (95%CI 23 ,58-2 4, 33)), ho spitaliza t ion s peak ed again in 49 Nove m ber 2020 (7 2,8 2 per 100.00 0 person year s (95% CI 7 2, 17-73, 48)). The standar diz ed 50 rat e rat ios (SRR) f o r AKI an d AKI-rela t ed m or tal ity with the o v er a ll A SR for CO V ID-19 51 hospitaliz at ions in t he de nominat or, decreased throu ghout the ob s er v at ion per iod and 52 rem a ined lower in au tumn than they were i n spring. In contr as t t o all CO V ID-19 53 hospitaliz at ions , the SRR for overall m ortality in C O V ID-19 ho spitaliz at i o ns diverged from 54 hospitaliz at ions with A KI in autumn 2 020. 55

56 Our stud y f or t he fir s t time pr ovid es nationwide data on COVID -19 r el at ed hospitaliz at ions 57 and acute kidney injury in Germany in 2020. AKI was a relevant compl ication and ass o ciated 58 with high mo r tali ty. W e observed a less prono unced increase in the A SR for AKI- rela ted 59 mor tali ty during aut umn 2020. The pr oport i on o f AK I-relat ed m o rtali t y in compar is on t o th e 60 overall m or tal ity decreas ed t hrougho ut the cour s e of t he pandem ic. 61 62 63 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 4

64 The SAR S-CoV-2 pandemic is a majo r challenge for patien t s , healthcare p rofess ion a ls , and 65 populat i o ns worldwide. Whil e ini t i al repo rt ing focused mainly on lung involv ement, th e 66 ongoing pandemic sho wed t hat m ultiple or gans ca n be inv o l ved, and p r ognosi s i s la r gel y 67 influenced by multi-organ involv ement. According t o publis hed dat a, the kidney is the 68 s e c o n d m o s t i n v o l v e d o r g a n . I n a s t u d y f r o m t h e N e w Y o r k C i t y a r e a , 2 2 , 2 % o f h o s p i t a l i z e d 69 patient s s h ow ed a cute k idn e y injury (AKI) (1). A m eta-analy s i s ( 2) confir med the h i gh AKI 70 r a t e a m o n g C O V I D - 1 9 p a t i e n t s . I n c o n t r a s t t o t h i s , t h e r e i s a l s o d a t a w i t h l o w e r i n c i d e n c e o f 71 AKI( 3,4) espe ci ally ear ly analysi s fr om Wuhan, Ch i n a did no t show high inc idence rates of 72 AKI (3–5) . H owever, diffe ren t def ini tions of AKI complica t e com parability. In crit ic ally il l 73 patient s w it h CO VID -19 a s man y a s 55 t o 90% exper i en c ed AKI ( 6–8), of whom 21 to 3 7 % 74 r e q u i r e d r e n a l r e p l a c e m e n t t h e r a p y ( R R T ) ( 6 , 8 , 9 ) . T h e m o r t a l i t y r a t e s o f p a t i e n t s w i t h a c u t e 75 kidney injury ar e si gnif i cant l y h igher than in patien t s with out A KI. In a study b y Chan et al . 76 (9), t he in-hospital mor ta lity was 50% amo ng patients with, ver su s 8 % amo ng those wi t hout 77 AKI. A n analysis b y Fisher et al. (8) fo und similar mo r tal ity rat es (52% vs. 1 9.6%). In addition, 78 the length o f s t ay ( L OS) wa s mar k ed ly pr olonged when AKI occ u rred ( 9,10). 79 The German healthcare s ys t em offer s compr ehens ive insurance cover ag e and , b y 80 inter nat ional c o mpar ison, a high le vel of heal t hcare s er v ices. Th rough out t he CO V ID-19 81 pandemic t here were no apparent ac ut e care r es our ce limitat ions, and the ho s pit a l 82 capa c it i es were at their li mit but no t overwhelmed. Nationwide un s electe d dat a , es pecially 83 covering a longer time int er val is n ot available. Ther e is one ob s er v at ional study, whic h 84 included 10.02 1 ho s p i t ali zations (11 ). The data fr om this stud y wa s ob t ai n ed f rom on e 85 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 5 German h e alt h insurance compan y a n d covered a s hor t period o f time (fr om Febr uary 26 86 until A pril 19, 2020). In this s t ud y, AKI was not addres s ed. 87 Our aim was t o obtain nationwide retrospective popu l at ion -based data on hospitaliz at ions 88 with CO V ID-19 and A KI. Our quer y c overed data fr o m the year 2020 an d contains , to our 89 knowledge, the most c om prehensive dat a regarding patients with COVID- 19 and AK I in 90 Germany. 91

& Methods 92 Data Source 93 In accor dan c e with the German Hospital Financ in g Act (KH G ) , a univer sal , per formance-94 based remun eration system was introd uc ed for gener al hospital servic es. The basis for t his 95 is the G er man-D i agnosis Related G r oups s ys tem (G-D RG system), where by ea ch inpatient 96 ca s e of tr eatment is r emuner a ted by m ea n s of a cor responding DR G rate. All hospitals 97 submit their hos p i t aliz ation dat a to the Institut e for t he hos pit al r em uner ation s ystem 98 (Institu t für das Ent gelts ystem im K ra nkenhaus G mb H, InEK). A fter s ubm is sion, a plausi b i lit y 99 contr ol of t he data is carried o ut by In EK. The s ub mission of hos pi t al ization d a t a is 100 mandat ory for r eimbursement of hospital s t ays. Thi s lead s to a stro ng ince nt i ve for hospitals 101 to supply complet e data, covering vir tu a lly every hospital in Germ a n y . Th e transfer red da ta 102 include i n for m a t ion on age, sex, discharge type, prim a r y a n d seconda ry ICD -10- GM(12) 103 coded diagno sis (In ternat ional Classifica t ion of D is ease s , 10t h Edition, Ger m a n Modification 104 Ver s ion 2020) and p e r for med oper ations and pr oc ed ures ( Operation en- und 105 Pro z edur enschlü ss el V ersion 2 020, O PS) (13). We per formed a query of G-D RG data for the 106 year 2020 via t he InEK data port al ( http s :// dat enbrowser.inek.or g/). T he InEK data are 107 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 6 avail able in ag gr egated for m, so that diffe r ent iation by federa l stat e is no t pos s ible. 108 Dif fer ent i ation by age can on l y be m ad e i n the pred e f ined age groups and differ en tiati on b y 109 gender only indirectly. The G-D RG data provided by InEK hav e been des c r ibed in det a i l 110 els ewher e (14). 111 Cohort selection 112 We incl u ded hospitaliz at ions w it h a sec o ndar y diagno si s of RT-PCR proven CO V ID-19 113 infection (ICD -10-GM: U07.1), aged over 15 ye ar s . We ex cluded hospi taliz at ions 15.525 114 (9,1%) with hos pit al t ransfer and un k nown dischar ge statu s fr om our a n al ysis. We identified 115 hospitaliz at ions w it h ac u te kidney injury by t heir corr es pon ding I CD-1 0-G M codes ( S1 116 Table). We i n c luded ho s pit a lizations wi t h ac u te kidney i n jur y s t ag e s 1 to 3 (A KIN 1 t o 3) (1 5). 117 We furt her s earched for OPS code s indicating hem odial ysi s and me chanical ve n tilation (S1 118 Table). The survival stat us at the end o f each ho spitaliz ation was obt a i n ed by dischar ge 119 type. The observation per iod was fr om 01.02.20 20 thr oughout 31.12.2 020 . 120 Statistical analysis 121 We c al culated crud e , a ge- s pe cific and age-s t andardized hospitaliz at ion rates (A S R ) pe r 122 100.000 per son years and the respe c t iv e 95% confidence int ervals (95% CI) . We u sed the 123 standard population of Germany in 2011 (16), provided by t he Federal Bureau of Statistics 124 (D ES TA TI S), f or direct a g e- s t andardi za t ion (S2 Table). We fur ther calc u lat ed standar diz ed 125 rat e ratios (SRR ), with t heir r es pe ctive confiden c e i n tervals . Continuou s var i ables are 126 repo rted with standar d devia tion (SD). Categor i cal data is repor ted in absolute number s and 127 percent a ges . We u s ed t he I CD-1 0 v er s ion o f the Elixhau s er comorbidit y in d ex (EC I) to a s s e s s 128 comorb i d i t ies ( 17). We excluded a cut e kidney injur y a s a comorb i d i t y . All data analysi s wa s 129 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 7 carried out usin g R ( Ver sion 3.6. 3) (1 8), fo r dir ec t age- s t andardization t he epitools pac kage 130 (19) was u s ed, cont i n g en cy tables wer e ev aluat ed us ing t he epiR pac ka ge (20). 131 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 8

132 General characteristics and age-standardized rates 133 In 2020, there wer e 16.776 .845 ho sp italizations in G er man hospitals. W e detected 154.170 134 hospitaliz at ions with RT-PCR pr oven COV ID-19 d i agno sis throu g h out the ob s er v at ion period . 135 There were 79.781 (51,8%) male and 74.382 (48,2 %) female pat i ent s hospitaliz ed, 136 respectively (Table 1). In 29.329 (19,0%) of the COVID- 19 a s s ociated hospitaliz at ions the 137 tr eatmen t included an int ensiv e ca r e unit (ICU) s t a y. The age-standar dized hos pit al ization 138 rat e f or CO V ID-19 in G er many was 2 32 ,8 per 100.0 00 person year s (95% CI 231,6-23 3,9) in 139 2020. M os t ho s p i t aliz ations (1 394, 1 p er 100.00 0 per s on year s ) o ccur red in th e ag e gr oup of 140 80 year s a n d older (Fig 1). 141 142 143 144 145 146 147 148 149 150 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 9 A KI ICU Ove r all N 16. 773 2 9.32 9 1 5 4. 1 7 0 Gende r % n % n % n ma l e 6 1, 5 ( 1 0.3 08) 62, 6 (1 8. 361 ) 5 1 , 8 ( 79. 781 ) fem a le 3 8,5 ( 6.4 6 3) 37, 4 (1 0. 965 ) 4 8 , 2 ( 74. 382 ) A ge gr oup ( y e a r s) % n % n % n 16 -1 7 0,0% 1 0,1% 38 0,3% 472 18 -2 9 0,4% 70 1,7% 49 3 4,0% 6143 30 -3 9 0,7% 112 2,8% 82 9 5,3% 8185 40 -4 9 1,9% 326 5,8% 17 04 7,5% 11601 50 -5 4 2,6% 443 5,8% 16 88 6,2% 9537 55 -5 9 4,5% 754 7,9% 23 12 7,5% 11602 60 -6 4 6,8% 1134 9,6% 28 18 7,7% 11858 65 -7 4 20,3% 3408 21 ,8 % 63 98 16,5% 25418 75 -7 9 17,0% 2851 14 ,6 % 42 82 11,7% 18045 80 + 45,8% 7674 29 ,9 % 87 67 33,3% 51309 le n g th of s ta y (d ays, SD) 1 8,4 ( 1 6 , 9) 1 8, 3 ( 1 6,7) 11, 3 ( 1 1,9) El ix hau ser c o mo r b i d ity ind e x ( %, n) % n % n % n 0 60 ( 1 0 . 0 41) 6 5 ( 1 8.9 1 8) 6 5 ( 9 9. 78 3) 1 – 4 24 ( 4 . 0 2 9 ) 2 5 ( 7. 18 6) 2 5 ( 3 9. 11 7) >=5 16 ( 2 . 7 0 3 ) 1 0 ( 3. 22 5) 1 0 ( 1 5. 27 0) me a n (S D) 2,0 ( 3,8) 1, 6 ( 3, 8 ) 1, 2 ( 3, 1) Table 1. G ener al char act e r is t ics of C O VID -19 hospitaliz ation s in Ge r man y 2 020. 151 ICU: intensiv e care unit - hos p i t aliz ati on s with inten s ive c ar e treat men t du r ing the hospital 152 sta y, AKI: a cute kidney i n j u ry - ho spitaliz at i o ns with acute kidney injury (A cut e k idney injur y 153 networ k St age 1- 3), SD: standard deviation 154 155 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 1 0 Fig 1. Age-spe c ific rat es of C OVID -19 hospitaliz at ions in Germany 2 020 (pe r 100.000 per son 156 years) 157 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 1 1 Hospital incidence of acute kidney injury 158 Acute kidney injur y, in cluding AKI Stage s 1-3, was dia gnose d i n 16.7 73 ( 10, 9%) of the 159 hospitaliz ed c a se s wi t h C OVID -19. Th e A S R for AKI was 25. 8 (95% CI 25. 4-26.2) per 100.00 0 160 person year s ( Table 2). W h i le men and wom en we r e equally d istr ibuted among t hos e 161 h o s p i t a l i z e d f o r C O V I D - 1 9 , m e n w e r e m o r e a f f e c t e d b y A K I t h a n w o m e n . T h e r e l a t i v e r i s k o f 162 AKI for m al es wa s 1,4 9 (95% CI 1, 4 4-1,53) compar ed to females hos pit al iz ed with C O V ID-19 163 (S2 Table). W e det ermined 10.310 (61,5%) m al e and 6.463 (38, 5%) female hospitaliz at ions 164 with acute k idn ey injury , respe ctively (Table 1). The propor tion of the se ver ity o f acute 165 kidney injury wa s d istr ibuted as follo ws: 32 ,0% (n= 5.40 7) developed A KI Stage 1, AKI St a ge 166 2 was present in 27,4 % ( n= 4.625) and in 40,6% (n= 6.842) AKI St ag e 3 was c o ded. There 167 were 101 hospitaliz ation s with more than one c od e fo r AKI. When acute k idn e y i n jury wa s 168 present du ring hospital s t a y, 8.3 74 (4 9,9 %) n e ed ed i n tensi ve care. Furt her mo re, mechanic al 169 ventilat ion was requir ed i n 7.0 50 (4 2, 0 %) ho spitaliz at i o ns with acute kidne y injur y (Table 2). 170 The relative r i sk for hos pit ali zations with AKI t o be t r eated in an int ensive c ar e unit was 3 ,27 171 (95%CI 3 ,21-3 ,3 4) (Table 3). 172 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 1 2 A K I I C U O v e r a l l n % n % n % Overall 1 6.773 (10 ,9%) 29.329 (19,0%) 154 .170 (- ) R e n a l re p l a c e m e n t therapy 3.4 43 (20,5%) 3.046 (10,4%) - males 2.6 35 76 ,5% 2.335 76,7% - - females 808 23 ,5% 711 23,3% - - I n - h o s p i ta l m o r ta l i ty 9.717 (57 ,9%) 12.322 (40,6%) 30. 300 (19,7 %) males 6.2 83 64 ,7% 7.990 64,8% 17.635 58, 2% females 3.4 34 35 ,3% 4.330 35,2% 12.665 41, 8% AKI Stage s 1 5.4 07 32 ,0% 1.865 20,3% - 2 4.6 25 27 ,4% 2.011 21,8% - 3 6.8 42 40 ,6% 5.333 57,9% - Chronic kidney disease 5.524 (32 ,9%) 5.365 (18,3%) 23. 380 (15,2 %) G3 3.6 38 21 ,7% 3.225 11,0% 15.270 9,9 % G4 1.4 90 8, 9% 1.084 3,7% 4.535 2,9 % G5 396 2, 4% 1.056 3,6% 3.575 2,3 % Mecha nical ventila t ion 7.050 (42 ,0%) 17.144 (58,5%) - males 5.0 52 71 ,6% 11.508 67,1% - females 1.9 98 28 ,3% 5.634 32,9% - A g e - s t an dar di z e d r at e (per 100.0 00 PYRs, 95% CI) 25, 8 (25,4-26 , 2) 44 ,0 (43,5-4 4 ,5) 232,8 (2 31,6-233,9 ) 173 Table 2: Results CO VID -19 ho spitaliz at ions in Germany 202 0, 174 ICU: intensive c are unit - hospitaliza tions with int e n s ive care t reatment d ur ing the ho s pit a l 175 sta y, AKI: a cute kidney injur y - hospitaliz at i o ns with ac ut e kidney injur y (acute kidney injur y 176 networ k, S t age 1- 3), 95 %C I: 95 % c on f i d enc e int erva l, PYRs: per s o n year s 177 178 179 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 1 3 180 ICU N on- IC U O v er a ll N % N % n AKI 8.374 49,9 8.3 99 5 0,1 16.773 Non- AKI 20.955 15,3% 116 .442 8 4,7% 137.397 Overall 29.329 19,0% 124 .841 8 1,0% 154.170 Dec ea s ed Alive O v erall N % N % N AKI 9.717 57,9% 7.0 56 4 2,1% 16.773 Non- AKI 20.583 15,0% 116 .814 8 5,0% 137.397 Overall 30.300 19,7% 123 .870 8 0,3% 154.170 181 Table 3: Contingenc y table for AKI an d risk of ICU a d mi ssion or in-hospital mor tali ty, 182 ICU: intensiv e care unit - hos p i t aliz ati on s with inten s ive c ar e treat men t du r ing the hospital 183 sta y, AKI: a cute kidney i n j u ry - ho spitaliz at i o ns with acute kidney injury (a c ute kidney injury 184 networ k, S t age 1- 3) 185 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 1 4 Renal replacement therapy 186 When acute kidney injur y was pr e s ent , renal replac em ent t herapy (R RT) was perfor med in 187 3.443 hos pit al izations, 20 ,5% of h os pit al izations with AKI and 2,2% of a ll C O V ID-19 188 asso ciated hospitalizations. Furth erm or e, when RRT was condu cted, m ec h anical ventilat ion 189 was need ed i n 2.380 (69,1 %) ca s e s. The modality of r enal replace ment t h erapy was 190 inter m ittent R RT in 230 ( 6,7% ) , and continuous o r p rolongend intermitt end RRT in 3.213 191 (93,3%) hospitaliz at i o ns , re spec t iv ely. In 828 (24, 0%) ho s pitalizations mor e t han on e 192 pro c edur e c o de i n dic at i n g RR T w a s present. 193 Chr onic kidney di s ea se (C KD St ages 3 to 5) was present in 23.380 ( 15. 2 % ) hospitaliz at ions 194 with A KI. We ob served 3.638 (21 ,7% ) hospitaliza t ions with AK I and CKD S tage G3, in 1.4 90 195 (8.9%) hospitalizations C KD G4 w a s pr evalent and 396 (2,4 % ) had CKD G5, r espec t iv ely. 196 In-hospital mortality 197 198 For all hos pit al izations w it h C O V ID-19, the in-ho s pital mortali ty amoun t e d t o 19 ,7% (n= 199 30.300). O f these, 58, 2% (n= 17.635) were m en and 41,8% (n=12.665) we r e women (Table 200 2), r es pectively. The relat i ve ris k fo r in-hospital mortality fo r men amo un t ed t o 1,30 (95%CI 201 1,27-1,3 3) comp ared to women ( S 3 Table). The A S R for in-ho s pital mo rtali ty wa s 4 7,7 202 (95%CI 47, 2- 48,3) per 100. 000 per so n year s. M o s t death s (51 1,2 per 100. 000 per s on year s ) 203 were observed in t he age g r oup over 8 0 year s of a ge ( F ig S1). When AK I was present, 9 .717 204 died, accounting fo r 57,9% of deaths in this grou p (Table 2), t he relati ve ris k f or in-ho s pital 205 mor tali ty was 3, 87 (95% CI 3,8 0-3,9 4) when AKI occurred (Table 3). M oreover, when 206 hemodialysis was perfor med, 2.891 died, which amo unts t o 61,4% of al l hospitaliz at ions 207 with RRT. 208 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 1 5 209 Length of stay 210 The mean length of s t ay (LOS) for hos pit ali zations with C OVID-1 9 was 1 1,3 da y s ( S D 1 1,9). 211 When acute kidney i n jury wa s code d, t he LO S incr ea sed to 18,4 day s ( SD 16,9) . For c ase s 212 that had a n IC U sta y the mean LO S w as 1 8, 3 d ay s (SD 16 ,7) ( Ta b l e 1) . 213 Time course of hospitalization rates and in-hospital mortality 214 215 The age- standar diz ed hos p i t aliz ation rates f or C O V ID -19 took a b i m odal course dur ing the 216 observation p eriod. The first peak of A SR occ u rred i n Apri l 202 0 (23,95 p er 100.000 per s on 217 years (95 %CI 23 ,58-24, 33) ), after th e ASR r os e steadily beginning in Feb ruar y . Sub sequent ly , 218 ASR for COVID-1 9 ho s pitalizations peak ed ag ain in N o v emb er 2020 (7 2,82 per 1 00.0 00 219 person years (95% CI 72, 17-73,4 8) ). In comparison to t he first peak , ASR ro s e steadily f or two 220 mont hs bef ore reaching the highest r at es (Fig 2A). 221 The A S R for AKI and AK I-re lated mo rt ali t y took a flatter cur v e, but o v erall fol low a s imilar 222 bimodal course. The ASR fo r AKI and AK I-relat ed m ortal ity r eac hed t heir peak in Mar c h 2 020 223 (for AK I 3,4 per 100.000 p erson years (95% CI 3,27-3, 55), which wa s al s o when AKI- rela ted 224 mor tali ty reached 2,07 (95 %CI 1 ,96- 2 ,18), respe ctively ) . In N ovember 2020 ASR tipp e d ag ain 225 when ASR fo r AKI stood at 8,07 per 100.000 per s o n year s (95 %CI 7,8 5-8 ,2 9), and the A SR for 226 AKI- relat ed mortality rose to 4,65 (95 % C I 4 ,4 9-4,8 2)) (Fig 2A). 227 The standard i zed r a t e ra tios (SRR) fo r A KI and AKI -re lated m or talit y wit h t he overa ll A S R f or 228 CO VID -19 hospitaliz at ions in the denom ina tor, ac t ual ly decr ea sed t hroughout the 229 observation period (Fig 2B) and r emained lower in autum n t han t hey wer e in s pr ing . In 230 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 1 6 contr as t to t his f i n ding, the A SR and in par ticular th e SRR for overall mort ality in C O V ID-19 231 hospitaliz at ions r os e sharply a gain as early as Sept ember 2020 (Fig 3A and 3B) . 232 The distr ibution of age grou p s var i ed thr oughout the observation period . In par tic u l ar , the 233 age group of 80 year s and above wer e most pr ev a lent in the m ont hs of Ap ril, November an d 234 December ( F ig S4). H ospitaliz at i ons in which AKI wa s c od ed, were pro p ort i o nat ely mu c h 235 older, with t he 80 years and abo v e ag e group b ei n g t he most repr es ent ed ( Fig S5) . 236 Fig 2: A) T im e c ou rse of age- standard i zed r ates (per 100.0 0 person years) of o v er al l 237 hospitaliz at ion, AKI and AKI-r el at ed mort a lit y for CO V ID-1 9 in Germany (202 0); B) 238 Standar diz ed r ate r a tios (SRR) of AKI age-s t andardized rat es ( per 100.0 0 person year s) and 239 AKI- relat ed mort al ity in relation t o the overall a ge- s tand ardize d COV ID -19 hos pit ali zation 240 rat e in German y (2 020) 241 Fig 3: A) T im e c ou rse of age- standard i zed r ates (per 100.0 0 person years) of o v er al l 242 hospitaliz at ion and over a ll m o rt a li t y for CO V ID-19 in G er man y (2020); B) Stand a r diz ed rate 243 rat i o s (SRR ) of ov erall mo rta lity age-standar diz ed r ates (per 100.00 p erson yea r s ) in relation 244 to t he ov erall a ge- s tand ardize d CO V I D-19 hospitalization rat e i n Germ a n y ( 2020) 245 Comorbidities 246 The Elixhaus er c o morbidit y index (EC I) varied a mong t he di ffe ren t group s . The mean EC I for 247 all CO V ID-19 a s s o ciated hospitalizations wa s 1, 2 (SD 3,1). For ho s pitalizations requiring 248 i n t e n s i v e c a r e t h e m e a n E C I w a s 1 , 6 ( S D 3 , 8 ) . W h e n a c u t e k i d n e y i n j u r y w a s p r e s e n t t h e E C I 249 increased t o 2 ,0 (SD 3 ,8) . In general, the four mos t comm on coded co mor bidi ties were 250 uncomplicated hyper tensi o n (35,3%), fluid and electr o ly t e dis or ders (20,6 %), uncomp l icated 251 diabetes mell itus (13,3%) and car dia c arr hythmias (10,0%) (Fig 4). When acute kidney injur y 252 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 1 7 or an IC U stay were coded, cardia c a r rhyth mias were r elative l y mo re pr eva lent comp a r ed t o 253 all hos p i t ali zations with COVID -19 (Fi g S 2 , S3). 254 Fig 4: Relative f requ encies o f c o morbidities in COVID- 19 ho s pit al izations in G er many 2 020. 255 chf: c on gesti ve heart failure; carit: c a rdiac arrhythmias ; v a l v: v a lvular d i s ea s e; pcd: p ulmonary circulati on 256 disorders ; p vd: perip h eral v asc ular d isorde rs; h ypunc : hypertensi on - uncomp lic a t e d; hypc : hypertensi on 257 - comp licated ; para: paral y s is ; ond: other ne urol og i c a l dis orders; c pd: c hronic p ul mona r y diseas e ; 258 d i a b u n c : d i a b e t e s - u n c o m p l i c a t e d ; d i a b c : d i a b e t e s - c o m p l i c a t e d ; h y p o t h y : h y p o t h y r o i d i s m ; l d : l i v e r 259 disea s e ; pud: pe ptic ulcer dis e ase- ex c l u din g ble e di n g ; aids : A I D S / HIV; l y mp h: l ympho ma ; metacanc: 260 meta s tatic c an c e r; s o li dtum: so l id tumou r: w i t h out m etast a s i s ; rheumd: rhe u mato id arthritis/ collaged 261 va s cula r dis e ase ; coag: coa g ul op athy ; obes: obe sity ; w loss : w e ight l o s s; fed: fluid a n d e le c trolyte 262 disorders ; b la ne : bloo d los s a n aem i a; da ne : de f ic i enc y a n aem i a; a l c ohol: a l c ohol ab u s e; drug : dru g a bu s e ; 263 psycho: psycho ses; depre: dep r e s s ion ; 264 265

266 Our s t udy f or the first tim e pr ov ides nationwide data on CO V ID-1 9 h ospitalizations and 267 acute kidne y injury in G erm any in 2 0 20. The ke y findings that emerge fr om our data ar e as 268 follows. The highes t pr oport ion of h ospitalizations as s o ciated wi t h C O V I D -19 were in th e 269 age gr oup over 80 years . The g ender distr ibut ion was equal in C OVID -19 related 270 hospitaliz at ion, however, we ob s erve d that m en mor e oft en deve l o ped ac ute kidney injur y 271 (RR 1,49 (95%CI 1, 44-1,53)), needed ICU c are and h a d a higher ris k of in- hospital m orta lit y 272 (RR: 1,30 (95% CI 1 ,27-1,3 3)) than women. O u r results thu s ex hibit a gender bias . Thi s i s 273 con s istent with a meta-analysis inc l uding 3.111 .714 C OVID- 19 a s s o ciated hospitaliz at ions 274 worldwide, which iden tified male sex as a ris k fa ctor for death and ICU-admis sion, d espite 275 the gender ratio being equal at the time of admis sion ( 21). Dif ferent under lying 276 mechanism s - from hor monal, genetic to behavioral ef fects and r i s k fa c t or prevalence – 277 have been dis cu s s ed in t he li t eratu r e (21,22). 278 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 1 8 Ov er all, 19 % of all h os pit a lizations needed intensiv e care tr eatment. This fraction wa s 279 part ic ular ly high in patient s with acute kidney injury. This finding is in line with other s t udies 280 (23,24) . Ther e are no cl ear ly defined criter ia fo r ICU admi s s ion in C O V I D-19 hospitaliz ed 281 patient s . Germany po s se sses t he h ighes t c apacit y of i n tensi ve care beds in all OE CD-282 countr ies . D es pit e t he pr esumably higher availa b i l it y of ICU capa city, this f ac t did not lead to 283 higher ut i lization of these c ap a cities. This could be an indicator t hat t he percent a ge of 284 critically ill patient s was similar compar ed to di fferent countr ies (23) . 285 O c cur ren ce of AKI con siderably differ s in previous studies, part ly du e t o a diff erent 286 definiti o n o f AKI. Our anal ysi s s howe d that 10.9% of all ho s pitalizatio ns experienced acute 287 kidney i n j u ry. The incidence of AKI in this nat i o nal series is lower than what has been 288 repo rted i n regionally confined stu die s f rom t h e Unit ed Stat es ( 8,9,23). A meta - a n a lysi s by 289 Ga b a r re et al. (2) r endered si m il ar r es ult s to our figur es . The y ob served an a ver ag e AKI 290 incidence of 11% (95% CI 8–1 7%) ov er all, wi t h highest ranges in the c r itically ill (23% (95%CI 291 14– 35%)). Nevert hel es s, the a ge s t ructur e, ris k f a c t or pr ev alence and th e burden of CO V ID-292 19 during the o bs er v at i on period di ffer relevan tly between count r ies and the AKI rate is 293 ther efo re on l y compar able t o a li mite d extend. 294 A lar ge p ropo r tion of pa t ie n ts with acute kidney inju ry suffered s evere ki d ney injury (A KIN 295 Stage 3: 40,6 %) , this is con si s t ent wi th prev ious findings (9). In the gr oup of h os pit al ization 296 with AKI 49 ,9% requir ed ICU adm is s ion compar ed to 15,2 % of patients witho ut A KI. The 297 relativ e r is k t o be tr ea ted in an ICU w as 3 ,27 f or AK I pat ients. M oreove r, i n th e gr oup wi th 298 AKI, 42, 0 % requir ed mechanical ve n tilati on. D ata fr om t he early phase of t he pandemic in 299 New Yor k state estimat ed that 53,6% of AKI pat i en ts needed in vasi ve ve nt il at i on (10) . This 300 higher num bers are lik ely ref lec t i ve o f m ore sev er e d is ease outcome s in patients with AKI. 301 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 1 9 RRT wa s r equired in 2,2% o f all hos pitalizations. Amo ng patien t s wi t h A KI 20, 5% r equired 302 RRT. Thi s figure i s c on sistent with d a t a from var i o us studies (6,9 ,25). The major ity (88,5%) of 303 the RRT wa s per formed in t he intens ive c are unit. 69, 1 % of all patient s r equiring dial ysi s 304 needed i n v asive ven tilat ion. Simi lar f i n dings w er e r eported in a mu lti -cent er cohor t stud y 305 fr om th e United S t a tes (6). Pr ev iou s studies sugge sted t hat t here is an asso cia t ion between 306 mechanical vent il at ion, AKI and RRT. R RT often occur s aroun d time of in t u bation, sug ge s t i n g 307 a role for a ltered hemodynamics in this situat ion ( 6,10). Hence, vasop r es sor s are often 308 initiat ed ar ound this time. 309 The over al l in-hospital m o rtal ity for CO V ID-19 a ssoci at ed hospitalizat ions vary a cros s 310 different count ries and coho rts, s h o w ing dif ferences i n t es t ing and case identification, 311 variable th r es holds for h os pit a lizati on and staff capacities . Ther efore, diff er ences in in-312 hospital mort a l it y r anged bet ween 1 5 t o 20% (26,27) . Our analy s i s sh owed that 30.300 313 (19.7%) o f hospitaliz ation s with C O V ID -19 ult ima tely decea sed, of 154.17 0 initia lly adm it ted 314 to hospital. In-hospital m orta lit y was 57. 9% among patients wit h AK I versus 15% amon g 315 those wi t hout AKI. This f i n ding under lines the poo r pr ognosis of these pat ients and the 316 sev er ity of disea se. 317 The age-standar diz ed hos p i t ali zation rate due to COVID- 19 in 2020 wa s bim odall y s haped, 318 with peak rat es i n Apr i l and N ovem b er. This o bs er v at ion is congruen t with the inci d enc e 319 figures r eport ed by th e Robert Koch In s t i t ute ( 28). The f irst peak with inc r eas in g m ortalit y 320 rat es in M ar ch and Apr i l wa s followed by a de cline in mor tality r ates a nd a concomitant 321 decrease in hospital admiss ions. Similar r es u l t s hav e a lr eady been d emons t rated in v a r ious 322 studies fr om ot her c o untries. In a dat a-analysi s fr om t he Unit ed S t ates a decr ea se i n the 323 risk-adju sted mor tal ity was ob served, ranging fr om 16.56 % to 9.29% in the early per iod of 324 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 2 0 this stud y (J anuary t hrough A pril 202 0) compared wit h the la t er pe r iod ( M ay thr ough June 325 2020) (2 9). Further , national ICU data fr om the Unit ed Kingdo m – pub lis hed a s a prepr int up 326 to now - conf i rmed thi s finding. The a uth ors found a redu c t i o n in m ort al it y r ates fro m 41,4 % 327 in March 20 20 to 24 ,8% in June 2 020 (30). The under lying reason fo r t he dec r eas e of 328 mor tali ty r a t es i s not obvious and cert ai n l y multifactor i al. O ne i m port a n t factor was 329 certainly th e c on comit ant decline in the incidence rate in t he UK and the U S. For ex am ple, a 330 large US s t udy of health i n s ur anc e d a t a, involv ing more t han 38.517 pati ent s , r eveal ed tha t 331 the s t rongest det erminant of i mprovements in in-hospital outcome wa s a d ecline in 332 communit y rates of inf ection (29) . Other reasons might be a s hif t i n dem ographics , 333 experience in physici an s t o early dia gnose and t reat C O V ID-19 patients, hospitals and staff 334 became les s overwhelmed dur i n g time. In our data during the s e cond peak in COV ID-19 335 hospitaliz at ions , whic h wa s more pronoun c ed than the f i r s t , th e in-hosp ital m orta lity rate 336 increased s t eadil y f rom Sept ember to December and exc eeded the CO VI D -19 asso ciated in-337 hospital mor tality in M arch and A pril mar kedly (5,36 per 10 0.000 person year s in A pril and 338 16,15 per 100 .0 00 p erson year s in D e c emb er). One c an on l y spe cula t e about th e f ac t ors that 339 led t o th i s devel opm ent. One, b ut cert ainly not the o nly, ex p l anat ion co uld b e a d ifferent 340 age dis t ribution of pat i ents adm itte d to t he hos pit al with a higher proport ion o f olde r 341 patient s in November and D ecember, as in our data ( F ig S 4, Fi g S 5) . F u rther , multiple factor s 342 might have influenced in- hos p i t al m or tali ty. The in creas in g number of patients in need of 343 critical care could hav e overwhelmed h os pit al s r e gi o nal ly, which happen ed fo r example in 344 Sax o ny, G er man y in D e cember 2020. A s alr eady mentioned, an Americ an s t udy ha s s hown 345 that hos pit al s did be tt er when t he pr eva lence o f COVID -19 in their s ur rounding 346 communit i es w as lower (29). Anot he r c ont ribut ing factor could hav e be en s t aff shor tages 347 driven b y CO V ID-19 infections, school lockdown s, quar a n tine regulat i o ns and the neces s it y 348 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 2 1 to t ake c ar e of relativ es infected with C O V ID-19 or in quarant ine. I n contr as t to the 349 developmen t dis cu s s ed above, th e ASR for AKI and AKI -rela ted m ortali ty had a 350 dispropor tionate ly smaller increase and were alr eady dec lining in D ecember. The a ge 351 distribut ion in our cohor t was s im ilar among all hospital ized patients , with a higher 352 pro porti o n of el d erly patient s i n the s e cond p eak in O c t ober, November and D e c em ber in 353 comparison to F eb ruary, M ar ch and Apr i l. Thu s , diff erences in age distr ib ution is not a good 354 explanation f or this finding (S4 Fig and S5) . At t his point , it must be s aid again th a t multip le 355 factor s could h a ve inf luenc ed the A S R for AKI und AKI-r ela ted mort ality. Ap art from 356 differences in th e p atient population or age distribut ion, change s in th e t reatm ent strategy - 357 e.g., th e early u se of dr ug ther apies f or pat ien ts with an incr eased r isk of s evere C O V ID-19 358 course s . 359 The mean length of s ta y (L OS - di scharged or dead) of all patients w as 11,3 da ys and 360 increased w it h AKI to 18,4 d a ys. LO S varies i n d i f feren t s t udies , and - according to a 361 s ystemat ic r ev iew by Rees et al. (31) - the median LOS was 14 da y s in Chin a, c o mpared with 362 5 (IQR 3–9) day s outside of China. Similar f igur es wer e r eport ed by Fi sher et al. (8). This m ay 363 be explained by many differences in c r iteria fo r ad miss ion and d i s char ge between count ries , 364 and d i f fe rent tim ings in the cour s e of t he pandemic. H ospital capaci t ies in G er many were 365 not o v er whe lmed by the pandem i c, and t he pr ess u re to dischar ge patients might have b een 366 lower. The m os t c om prehensive a nalysis of G er man patient s t o dat e included 10.021 367 patient s with a mean L OS of 1 4, 3 d a ys (32). This w a s longer than i n our analy si s , an 368 explanation fo r the dis cr epancy co uld be timing of data collec t ion. Karagiannidis et a l. 369 collec t ed their data at t he beginning o f th e pandem ic. D ue to an increasing experience in 370 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 2 2 tr eating C O V ID-19 patient s and new tr eatmen t op t ions, t he lengt h o f stay c onsequent ly 371 declined. 372 The ob s er ve d comor bidities varied between the differ ent gr oups. H yper tension, dia b etes 373 melli tus, fluid and e lectrolyt e dis o rd ers and c ar diac arrhyt hmias w er e the most commo nly 374 coded comor bidities . Th i s finding is plausible, as t hes e ar e the mo s t c om mon comor bidi ties 375 in patients with chronic kidney diseas e (3 3) and l ikewi se kno wn risk fac t or s for 376 hospitaliz at ion wit h C O V ID-19 (34) . Car dia c arr h yt hmias be came more pr ev alen t in 377 hospitaliz at ions w it h AKI. Cardiac arrh yt hmias are a common f inding in COVID -19, in 378 part ic ular w it h an i n c r eas ing number of comorb i d i ties (35) . 379 Str engths and limit ations 380 The major s t rength of th i s study is th at we provide a nat i o nwi d e popula ti on- bas ed v iew on 381 CO VID -19 ho s pit al izations with a c ut e k idn ey injur y in Germany . W e re t rieved DRG-data 382 covering virt ually every hospital in Germany. In th i s stud y, we u se r outine h os pit al ization 383 data, which is lac king c linical detail. T his might lead t o confound i n g t hat we could not adjust 384 for . The InEK provides aggre gated data du e to da ta safety pr ec aut ions , whi ch is f ully 385 anonymized. It is , th erefor e, possi b l e that th i s c ould lead t o bias cause d by readmis s ion, 386 thou gh this pos sibil it y i s limited, b ec ause of the s ever i t y of most ca s e s a nd our efforts to 387 exc lud e readmis s ions fr om the dat aset . Another imp or tant f ac t to consid er, when anal yzi n g 388 rou tine dat a , is the ob s er v at i o n peri od . In ou r analysis t he observation period ends in the 389 midst of the second wave of the CO VID -19 pandem i c in G er many. Thi s w ill lead t o bias, as 390 the outcome of thes e c a se s remains unknown, i n part i cular for hospitaliz at i o ns i n the la st 391 mont h of the observation p eriod. 392 393 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 2 3 394 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 2 4

395 1. R i char ds on S , Hirsch JS, N aras i mh a n M, Cr a wford J M, M c Ginn T, D a v idson K W, et a l. 396 Presenting Chara cter i stic s , Com orbidities, a n d Outcome s among 57 0 0 Pat i en ts 397 H o s p i t a l i z e d w i t h C O V I D - 1 9 i n t h e N e w Y o r k C i t y A r e a . J A M A - J o u r n a l o f t h e 398 Ameri can M e d i cal As s ociation. 202 0; 323(20):205 2–9. 399 2. Gabarr e P, Du ma s G, D upont T, Darmo n M, A z oulay E , Zafr ani L . A cute kidney injur y 400 in critically il l patients with COV ID-19. Intensive C are M edici n e [Inter net ]. 401 2020;46(7): 13 39–4 8. Avai lable f r om: ht tps:/ /doi.org/1 0.100 7/s001 34-020 - 06153-9 402 3. Wang L , Li X, Chen H , Yan S, L i D , L i Y, et al. Cor onav irus dis ease 19 infection does not 403

in ac u te kidney i n jury: A n analys is of 116 ho s p i t ali zed patients f rom W uhan, 404 China. Amer i can Jou rnal of Nephrolo gy. 2020 ;51( 5):34 3–8. 405 4. Yang X, Jin Y, Li R, Z h a n g Z, Sun R, C hen D . Pr ev alence a n d imp act of acut e renal 406 impairmen t on COVID -19: A s yste mat ic r eview and meta- a n a lysis. C ritical Care. 407 2020;24(1): 1–8 . 408 5. Zhou F, Yu T, D u R, Fan G, Liu Y, Li u Z, et al. Clinical cour s e a nd r is k factor s for 409 mort al i t y of adu lt inpatien t s wit h COVID- 19 in Wuhan, China: a retrospect iv e c oh ort 410 s t udy. The Lancet [Inter net]. 2020;39 5( 10229): 1054–62 . Available f r om: 411 http: / /dx.doi.org/10.10 16/S0140- 67 3 6(20)30566-3 412 6. Gupta S, Co ca S G, Chan L , M el am ed ML, Brenn er SK, Hayek S S, et al . AKI tr eated with 413 renal replacement t h erapy in critic ally ill patient s with COVID -19 . Jou rnal of the 414 Ameri can Society of Nephrology. 2 02 1;32(1) :161–7 6. 415 7. Thakkar J, C hand S, Aboodi M S , Gone AR , Alahi ri E, Sc h ec t er DE, et al . C haracteristic s , 416 O ut come s and 60- Day H os p i t al Mo r tality of IC U Patients with COVID -19 and Acute 417 Kidney Injury . Kidney3 60. 20 20;1(12) : 1339–44. 418 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 2 5 8. Fisher M , Neugarten J, Bellin E, Yunes M , Stahl L, Johns TS , et al. AKI in H ospitaliz ed 419 Patients with and with ou t CO V ID-19 : A C omparison St udy. Jou rnal of th e Ame r ic an 420 Soc iet y of N ephr olog y [Int ernet] . 2020 Se p 1;31(9) : 2145 LP – 21 57. A v ai lable f rom: 421 http: / /jas n.a snjourn a ls .org/ content/31/9/ 2 145.ab str ac t 422 9. Chan L, Ch audhar y K, Saha A, Chauhan K, Vaid A , Zhao S, et al. AKI in hospitaliz ed 423 patients with COVID -19. J ournal of t he A mer ican Society of Nephr ology. 424 2021;32(1): 15 1–60. 425 10. H irsch J S, Ng J H, Ro s s D W, Sh arma P, Shah H H, Bar nett RL , e t al. Acute kidne y injur y 426 in patients hospitaliz ed with COVID -1 9. Kidney Inter national. 2020;98(1 ):2 09–18. 427 11. Karagiannidis C , Mo ster t C, H entsch ker C , V o s haar T, Malzahn J, Schillinger G , et al . 428 Cas e character is tic s , re source u s e, an d ou tcomes o f 10†ˆ 02 1 pat ients wit h C O V ID-19 429 admitted to 920 G er man hospitals: an observational stu dy. The L an cet Res p i r ator y 430 Medicine [Int ernet ] . 2020 [cited 2021 Mar 11] ;8(9):853–6 2. Avai lable f rom: 431 ww w.t helanc et . c om / r es pir a t oryVo l 432 12. Graubner B, Auhub er TC. I CD- 10-GM Ver sion 2020, S ys t emati s c hes Ver z eichni s , 433 Intern a t iona le statistisc h e Klassifikation der Krankheiten und v er wandter 434 G e sundheitspro bleme, 10. Revis ion, S t and: 20. Septem ber 2019 [In te r net ]. Köln: 435 Bundesins t itut für Arzneim itt e l und M ediz inp rodukte (BfA rM ) i m A uft rag des 436 Bundesminis t eriums fü r Gesundheit ( BMG) u nter Be teil igung der Ar bei tsgruppe ICD 437 des Kurat oriums fü r Fr a gen der Klas sifikation im G esundheit s wesen (K K G ); 2019 . p. 438 Se p tember 2019. Available fro m : www.dimdi.de – Kla s sifikationen – D ownloads – 439 ICD-1 0-G M – Ver sion 2021 440 13. O per ationen- und Pr oz edu re ns c h l ü ssel Inter nationale Klass ifikation der P rozeduren 441 in der Med i zin (OP S) Band 1: S ys t ematisc h es Ver z eichni s [ Intern et]. Bund e sinstitut für 442 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 2 6 Arzneimit t el und Med i zinpr o dukte ( BfA rM ) im Au ft rag des Bundesministeriums für 443 G e sundheit ( BMG) unt er Be t e il igung der Arbeits gr uppe ICD des Kurator iums für 444 Fragen der Klassifikation im Ges un d hei t s wesen (K KG); 2020. 740. Available f ro m: 445 https:// w ww.dimdi.de/dynami c /de/ klas s if i kat i o nen/ icd/i cd- 10-gm/ 446 14. Bus se R , Nimpts ch U . CO VI D- 19-Pandemie: H i stor is ch niedrige Bet tenausl astun g . 447 Deut s che s A erzteblatt Online. 2021;1 18( 10):10–3. 448 15. L evin A, Warn ock D G , M ehta RL, Kellum J A, Shah S v. , M oli t oris BA, et al. Impr ov ing 449 O ut come s From A c ut e K i d ney Injur y: Report of an Initia t iv e. Am er ican J ournal of 450 Kidney Disease s. 2 0 07;50(1): 1–4. 451 16. G esund heitsberichter s t attun g de s Bunde s [Int ernet] . 2021. A v ailable f ro m: 452 https:// w ww. gbe-bund.d e/ gbe/p kg _i s gbe5.prc_i sg b e 453 17. Quan H , Sundar arajan V , H al f on P, Fong A. C oding algorit hms for defining 454 c om orbidities i n . M edic al Care [Inter net ]. 2005;4 3(11) . Available f rom: 455 http: / /scholar.goo g le. com/ s cholar ? h l =en & bt nG =Search &q=intit le:Coding+Algorit h ms456 +for+D e fining+Comor bi diti es+in#1%5Cnhtt p:/ /s c holar .goo gle.c om /s c ho l ar ?hl=en& b t457 nG =Sear ch&q=intit le : Coding+algorith m s +fo r+defin ing+comor biditi es +in#1 458 18. Team R C . R: A Langua ge and Env ir onment f or Stat i stical Comput ing [Inter net ]. 459 Vienna, Austr ia : R F o undation for Sta tistical Computing; 20 20 . Avai lable from: www.r-460 project.or g 461 19. Tom a s J. Ar agon. epitools: Epi demiology Tools . R pa cka ge v er s ion 0 .5- 10.1. 20 20. 462 20. Stevenson M, Ser g eant E. epi R - Tools f or the A nal ysis of Epidemiologic al D a t a. 2021 . 463 21. Pe c kham H, de Gru i jt er NM, R a ine C, Rad ziszew ska A, Ciurt i n C, Wedderbur n LR, et 464 al . M ale s ex iden tifi ed b y global COVID -19 met a - ana lysi s as a ris k f ac t or fo r death and 465 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 2 7 ITU admission. N atur e Comm unic at i ons [Int ernet]. 2020;11(1 ):1–10 . A v ailable f rom: 466 http: / /dx.doi.org/10.10 38/s4 1467-02 0- 19741-6 467 22. Kelada M, Anto A, D av e K, Saleh SN. The Role of Sex in t he Risk of M ort a l it y From 468 COV ID-19 A mongst Adult P a t ients: A Sys tem atic R ev iew. Cur eus . 2 020;1 2( 8). 469 23. R i char ds on S , H irs c h JS, N ar a simhan M , Crawfor d JM , McG inn T, Davids on KW, et a l. 470 Presenting Chara cteristic s, Como rbidities, and Out comes A mong 570 0 Pat i en ts 471 Hospitaliz ed W it h COVID- 19 in t h e N ew Yor k City A rea. JA M A [Inter n et]. 472 2020;32 3:205 2–9. Available fr om : htt ps:/ / jamanet w o rk. com/ 473 24. Fisher M , Neuga r ten J , Bel lin E, Yune s M, Stahl L, Jo hns TS, et al. A KI in H ospital iz ed 474 Patients with and with ou t CO V ID-19 : A C omparison St udy. Jou rnal of th e Ame r ic an 475 Soc iet y of N ephr olog y [Int ernet] . 2020 Se p 1;31(9) : 2145 LP – 21 57. A v ai lable f rom: 476 http: / /jas n.a snjourn a ls .org/ content/31/9/ 2 145.ab str ac t 477 25. R i char ds -B ell e A , O r z e chow s ka I, D oidge J, Thomas K, Harr is on DA, Koelewyn A, et a l. 478 Critical care ou tcomes , for the f i rst 200 pat ients with confirmed C OVID -19 , in 479 Engla n d, Wale s and Norther n Ireland: A repo rt from the ICNARC Ca se M ix 480 Progr a m me. Jour nal of the Intensive Care Society. 2 020;0(0): 1–10. 481 26. Wiersin ga W J, R hode s A , C heng AC, Pea coc k SJ , Pres c ot t HC . Pathophysiolog y , 482 Transmi ssion, D iagno s i s, and Treatmen t of Cor onavirus D i sea se 2019 ( C O V ID-19) : A 483 Revi ew. J AM A - Journ a l of the Ame r ic an M edic al Asso ci at ion. 2020;3 24(8): 78 2–93. 484 27. Sorci G , Faiv r e B, Mor and S. Explaining amo ng-countr y variation in COVID -19 ca s e 485 fatality rat e . Scienti fic Repor ts. 2020 Dec 1;10(1). 486 28. R obert K oc h Ins t i t ut. Tot al ov er view o f c ases, death s, and 7-da y incidenc e s repo rted 487 per day to the RK I b y state and count y. [In ternet ]. 2021. Available f rom: 488 https:// w ww.r ki.de/DE/ Content / Inf A Z/N/Neuart ige s _Cor onavirus/ D aten/ F allzahlen_489 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 2 8 Daten.h tml;jsessionid=4D411D FBB2E 2575A 9CD1 F5504 E7 CE3BC .i n ternet 0 52?nn=134490 90888 491 29. Asc h DA, Sheil s N E, I slam MN, Ch en Y , Werner RM, Buresh J , et al. V ariation in U S 492 Hospital Mor tal ity Rates for Patients Admitt ed with COVID -19 during the Fir s t 6 493 Mont hs of t he Pandemic. JAM A In ternal Medicine. 2020;191 04(4):4 71–8 . 494 30. Dennis JM , Mc Gov er n AP, Vollmer S J , Mateen B A. Improv ing C O V I D -19 crit ic al care 495 mort al i t y over t i m e in England: A national cohor t study, M arch to June 2020. 496 medRxiv. 20 20;44(0). 497 31. Rees EM, Nightingale ES, J afari Y, W at erlow N R , C l if ford S, Pearson CA B, et al. CO V ID-498 19 length of hospital stay: A sy stematic review and data synthe si s. medRxiv. 2020; 499 32. Karagiannidis C , Mo ster t C, H entsch ker C , V o s haar T, Malzahn J, Schillinger G , et al . 500 Cas e c har ac t eristics , r e s ou rce u se , a nd o utcomes of 10 021 p a t ients wit h C O V ID-19 501 admitted to 920 G er man hospitals: an observational stu dy. The L an cet Res p i r ator y 502 medici n e [Int er net ]. 2 020 Sep [cite d 2021 M ar 11] ;8(9):853–62. A va il able f r om: 503 ww w.t helanc et . c om / r es pir a t oryVo l 504 33. Reichel H, Zee J, Tu C, Y o un g E , Pisoni R L , S t engel B, et al. C hronic kidne y di sea s e 505 progr ession and m or tali t y risk profil es in Germ any: Results from the Chr on i c K i d ney 506 Diseas e Outcomes and Practice Patt er ns Stud y. Nephrolog y D ialysi s Transplantation. 507 2020 Ma y 1;35(5): 803 –10. 508 34. Ioannou G N, L o cke E, Green P, Ber ry K , O’H a r e AM , Shah J A, et al. Ris k Factor s for 509 Hospitaliz at ion, M ec h a n i cal V entilati on, or Death A mo ng 10Ei88131 US Veterans W ith 510 SARS-CoV -2 Infection. JA M A net work open. 2020;3(9 ):e202 2310. 511 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 2 9 35. Coromilas EJ, Ko c hav S , G oldenthal I, B iv iano A, Garan H , Goldbarg S, et al . 512 W or ldwide Survey of C O V ID-19 As s o ciated A rrh y t hmias . Circulation: Arr hythmia and 513 Elec t rophysiology . 2021; (Mar ch): 28 5 –95. 514 515 516 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 3 0 Supporting information 517 S1 Fig . A ge- s pecific mort ali t y rates of CO VID- 19 hos p i t ali zations in G er many 202 0 (per 518 100.000 per s on year s) 519 S2 Fig. Relative freq u enc ies of com orbidi ties in COVID -19 hospitalizat ions w it h AKI in 520 Germany 2 020. 521 chf: c on gesti ve heart failure; carit: c a rdiac arrhythmias ; v a l v: v a lvular d i s ea s e; pcd: p ulmonary circulati on 522 disorders ; p vd: perip h eral v asc ular d isorde rs; h ypunc : hypertensi on - uncomp lic a t e d; hypc : hypertensi on 523 - comp licated ; para: paral y s is ; ond: other ne urol og i c a l dis orders; c pd: c hronic p ul mona r y diseas e; 524 d i a b u n c : d i a b e t e s - u n c o m p l i c a t e d ; d i a b c : d i a b e t e s - c o m p l i c a t e d ; h y p o t h y : h y p o t h y r o i d i s m ; l d : l i v e r 525 disea s e ; pud: pe ptic ulcer dis e ase- ex c l u din g ble e di n g ; aids : A I D S / HIV; l y mp h: l ympho ma ; metacanc: 526 meta s tatic c an c e r; s o li dtum: so l id tumou r: w i t h out m etast a s i s ; rheumd: rhe u mato id arthritis/ collaged 527 va s cula r dis e ase ; coag: coa g ul op athy ; obes: obe sity ; w loss : w e ight l o s s; fed: fluid a n d e le c trolyte 528 disorders ; b la ne : bloo d los s a n aem i a; da ne : de f ic i enc y a n aem i a; a l c ohol: a l c ohol ab u s e; drug : d r ug a b us e ; 529 psycho: psycho ses; depre: dep r e s s ion ; 530 531 S3 Fig. Relative frequencies of comorbidi t ies in C O V ID-19 hospitaliz at i o ns with ICU 532 tr eatmen t in Germany 2020. 533 chf: c on gesti ve heart failure; carit: c a rdiac arrhythmias ; v a l v: v a lvular d i s ea s e; pcd: p ulmonary circulati on 534 disorders ; p vd: perip h eral v asc ular d isorde rs; h ypunc : hypertensi on - uncomp lic a t e d; hypc : hypertensi on 535 - comp licated ; para: paral y s is ; ond: other ne urol og i c a l dis orders; c pd: c hronic p ul mona r y diseas e ; 536 d i a b u n c : d i a b e t e s - u n c o m p l i c a t e d ; d i a b c : d i a b e t e s - c o m p l i c a t e d ; h y p o t h y : h y p o t h y r o i d i s m ; l d : l i v e r 537 disea s e ; pud: pe ptic ulcer dis e ase- ex c l u din g ble e di n g ; aids : A I D S / HIV; l y mp h: l ympho ma ; metacanc: 538 meta s tatic c an c e r; s o li dtum: so l id tumou r: w i t h out m etast a s i s ; rheumd: rhe u mato id arthritis/ collaged 539 va s cula r dis e ase ; coag: coa g ul op athy ; obes: obe sity ; w loss : w e ight l o s s; fed: fluid a n d e le c trolyte 540 disorders ; b la ne : bloo d los s a n aem i a; da ne : de f ic i enc y a n aem i a; a l c ohol: a l c ohol ab u s e; drug : dru g a bu s e ; 541 psycho: psycho ses; depre: dep r e s s ion ; 542 543 S4 Fig . Per centage distribu tion of age gr oups during th e ob s er va t ion period (Ov er all CO V ID-544 19) 545 S5 Fig . Per c en tage di s tr ibution of age gr oups d uring the observation per iod (AKI) 546 S1 Table. L ist of ICD -10- GM and O PS Codes, ver s ion 2020 547 S2 Table. Contingen cy table AKI by sex 548 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint 3 1 S3 Table. Contingen cy table in-hospital mor tal ity by s ex 549 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 4, 2021. ; https://doi.org/10.1101/2021.04.30.21256331doi: medRxiv preprint