{"paper_id":"da97df39-bc6d-471f-babc-71b988f9b23d","body_text":"The impact of positive peritoneal cytology on\nprognosis in patients with cervical cancer:\na meta-analysis\nSang-Hee Yoon 1, Soo-Nyung Kim *,1, Seung-Hyuk Shim 1, Ji-Young Lee 1, Sun-Joo Lee 1, In-Kyeong Oh 1,\nHyeon-Jeong Kim 1 and Soon-Beom Kang 1\n1Department of Obstetrics and Gynecology, Research Institute of Medical Science, Konkuk University School of Medicine, Seoul,\nKorea\nBackground: The impact of positive peritoneal cytology on the prognosis of cervical cancer is controversial. Thus, we performed a\nmeta-analysis to determine its impact on recurrence, and to investigate correlations between abnormal cytology and/or lymph\nnode metastasis in cervical cancer.\nMethods: A systematic literature review was conducted through July 2014. Odds ratios (ORs) and their 95% confidence intervals\n(95% CIs) were calculated by standard meta-analysis techniques with the fixed-effects models, if there was no significant statistical\nheterogeneity across studies by using I\n2.\nResults: Of 303 studies retrieved, 6 were included in the meta-analysis. These six case–control observational studies included 1360\ncervical cancer patients who showed negative peritoneal cytology and 64 who showed positive peritoneal cytology. Over the\ncombined study period, 20 of 45 in the positive peritoneal cytology group experienced recurrence, whereas 88 of 539 controls did.\nThe meta-analysis based on the fixed-effects model indicated a significant increase in the risk of recurrence in the positive\nperitoneal cytology group relative to the control group (OR: 4.47; 95% CI: 2.33–8.58, Po0.001, I\n2 ¼ 0.0%). Moreover, the results of\nour meta-analysis suggested that the positive peritoneal cytology group displayed more lymph node metastasis than the negative\nperitoneal cytology group (OR: 3.73; 95% CI: 2.13–6.53, Po0.001, I\n2 ¼ 0.0%).\nConclusions: Although based mainly on retrospective observational studies, our meta-analysis indicates that abnormal peritoneal\ncytology may be strongly associated with poor prognosis in patients with cervical cancer. Future research should verify this\nrelationship through prospective observational studies over a longer term.\nPeritoneal cytology examination is commonly performed for\nsamples obtained during gynecology surgery. The prognostic value\nof peritoneal cytology in ovarian cancer among gynaecological\nneoplasms is widely accepted, and it is included in the\nInternational Federation of Gynecology and Obstetrics (FIGO)\nnomenclature (1994) (Benedet et al , 2003). In endometrial\ncarcinoma, peritoneal cytology is included in the FIGO staging\nsystem, and positive cytology must be reported separately without\nchanging the stage. A positive cytology result is considered a\npoorer prognostic factor. However, only a few reports have\naddressed the issue of positive peritoneal cytology in cervical\nneoplasms (Benedet et al, 2003; Kuji et al, 2014). In cervical cancer,\nthe incidence of positive peritoneal cytology has been reported to\nbe 0–15% (Kilgore et al, 1984; Delgado et al, 1989; Takeshima et al,\n1997; Estape et al, 1998; Kasamatsu et al, 2009). It is reported that\nthe rate of positive peritoneal cytology among patients with\nsquamous cell carcinoma (SCC) in FIGO stage I or II disease\nis low (0.3–1.8%) (Delgado et al , 1989; Takeshima et al , 1997;\nEstape et al, 1998). On the other hand, positive peritoneal cytology\nwas found four times more frequently in adenocarcinoma (ADC)\nthan in SCC (Imachi et al, 1987). However, the prognostic value of\nperitoneal cytology in cervical carcinoma remains unanswered.\n*Correspondence: Professor S-N Kim; E-mail: snkim@kuh.ac.kr\nReceived 17 February 2015; revised 12 June 2015; accepted 25 June 2015; published online 30 July 2015\n& 2015 Cancer Research UK. All rights reserved 0007 – 0920/15\nFULL PAPER\nKeywords: cervical cancer; peritoneal cytology; prognosis; meta-analysis\nBritish Journal of Cancer (2015) 113, 595–602 | doi: 10.1038/bjc.2015.266\nwww.bjcancer.com | DOI:10.1038/bjc.2015.266 595\n\nPrevious reports have been inconsistent; whereas one study failed\nto show any prognostic inference of positive peritoneal cytology in\ncervical cancer (Abu-Ghazaleh et al , 1984; Delgado et al , 1989;\nTrelford et al, 1995; Takeshima et al, 1997), other studies reported\nthat patients with positive peritoneal cytology have a poor\nprognosis than those with negative cytology (Imachi et al , 1987;\nIto and Noda, 1992; Kashimura et al, 1997; Estape et al, 1998; Zuna\net al , 2009). Furthermore, some studies reported that positive\nperitoneal cytology was associated with poor prognosis only in the\npatients with ADC or adenosquamous carcinoma, but not in those\nwith SCC (Kasamatsu et al , 2009; Kuji et al , 2014).\nTherefore, we performed a meta-analysis using relevant studies\nto investigate the impact of positive peritoneal cytology on\nprognosis, focusing on lymph node metastasis and experienced\nrecurrence, in cervical carcinoma when compared with negative\nperitoneal cytology.\nMATERIALS AND METHODS\nSearch strategy. Two of the authors of the present study (S-HY\nand S-HS) designed the protocol and extraction forms in\naccordance with the Preferred Reporting Items for systematic\nReview and Meta-analyses (PRISMA) guideline. For this meta-\nanalysis, we searched online abstracts from PubMed, Embase, and\nthe Cochrane Central Register of Controlled Trials (CENTRAL) in\nthe Cochrane Library through July 2014. For this search, we used\nthe following keywords: ‘cervical neoplasm or cancer or carcinoma\nor malignancy,’ ‘uterine cervical neoplasm or cancer or carcinoma\nor malignancy,’ ‘carcinoma of the cervix or the uterine cervix,’\n‘invasive carcinoma of the cervix or the uterine cervix,’ ‘squamous\ncell carcinoma of the cervix or the uterine cervix,’ ‘adenocarcinoma\nof the cervix or the uterine cervix,’ and ‘positive intra-peritoneal\ncytology or positive peritoneal washing cytology or abnormal\ncytology.’ The titles and abstracts were checked to exclude any\nclearly unrelated articles. The full text of the remaining papers was\nevaluated to determine their relevance. In addition, the references\ncited in the selected papers and published reviews were checked to\nevaluate whether they included any additional studies of relevance.\nSelection criteria. Studies were included if (1) they were\nrandomised controlled trials, a prospective or retrospective cohort\nstudy, a population-based case–control study, or a nested case–\ncontrol study; (2) the participants of interest received surgical\ntreatment for cervical cancer; (3) the intervention of interest was\nperitoneal cytology; (4) the outcome measured was cervical cancer\nrecurrence and/or lymph node metastasis, measured as the relative\nrisk, the odds ratio (OR), or the hazard ratio estimated with the\n95% confidence interval (CI) (or sufficient data for calculating\nthem). Single-arm cohort studies were not included in the meta-\nanalysis. The studies that included advanced stage disease, such as\nstage III or IV, were excluded from the meta-analysis.\nData extraction. Data from each included study were system-\natically extracted by two authors using a standardised form. The\nform documented the most relevant items, including the name of\nthe first author, the year of publication, the study design, the study\nperiod, age, sample size (cases and controls or cohort size), tumour\nstage, histological type, peritoneal cytology positivity, follow-up\nperiod, and prognostic value (recurrence, lymph node metastasis)\nbetween positive and negative peritoneal cytology.\nQuality assessment. Study quality was evaluated independently by\ntwo authors using the nine-star Newcastle–Ottawa scale (NOS)\ncriteria (Stang, 2010). The NOS criteria included three categories:\n(1) selection: 0–4; (2) comparability: 0–2; (3) exposure (case–\ncontrol studies) or outcomes (cohort studies): 0–3. NOS scores\nranged from 0 to 9, with a score X7 indicating good quality. Any\ndisagreement was resolved after a discussion and reevaluation with\nthe third author.\nStatistical analysis. From the original study data, the OR and the\n95% CI were calculated for each study for the recurrence rate for\npositive and negative peritoneal cytology (Higgins and Thompson,\n2002). Heterogeneity across studies was examined using I2, which\nmeasures the percentage of the total variation across studies. Here\nsubstantial heterogeneity was defined as an I2-value 450%\n(Higgins et al , 2003). In the absence of significant heterogeneity,\na fixed-effects model was used. When there was statistical\nheterogeneity, a random-effects model was used to estimate the\ncombined OR for randomised and observational studies. Then, a\nsubgroup analysis was conducted for the type of histology (SCC,\nADC, or all type). The subgroup analysis was planned a priori\nbefore the data were collected and analysed. To evaluate the\ninfluence of single studies on the overall estimate, a sensitivity\nanalysis was performed. Publication bias was evaluated using a\ngraphical method. A funnel plot was built to assess this bias\nby using the s.e. and the OR (Sterne and Egger, 2001;\nSterne et al , 2001). Publication bias was evaluated using the\nBegg–Mazumdar rank correlation test, Egger’s test and the fail-safe\nN-test (Orwin and Boruch, 1983; Begg and Mazumdar, 1994).\nComprehensive Meta-Analysis version 2.0 (Biostat, Englewood, NJ,\nUSA) was used for all statistical tests. Po0.05 was considered\nsignificant for this meta-analysis. Data from this meta-analysis are\npresented according to the checklist based on the Meta-analysis Of\nObservational Studies in Epidemiology (Stroup et al , 2000).\nRESULTS\nLiterature search. Figure 1 presents a flow diagram showing how\nrelevant studies were identified. A total of 22 potentially relevant\npapers were found by focusing on abnormal peritoneal cytology\nand prognosis, particularly recurrence or lymph node metastasis\n(Creasman and Rutledge, 1971; Keettel et al , 1974; Hughes et al ,\n1980; Abu-Ghazaleh et al, 1984; Kilgore et al, 1984; Ziselman et al,\n1984; Willett, 1985; Roberts et al, 1986; Imachi et al, 1987; Delgado\net al, 1989; Zuna et al, 1990; Ito and Noda, 1992; Morris et al, 1992;\nPatsner, 1992; Trelford et al , 1995; Zuna and Behrens, 1996;\nZuna, 1996; Kashimura et al , 1997; Takeshima et al , 1997; Estape\net al, 1998; Kasamatsu et al, 2009; Kuji et al, 2014). Among these 22\npapers, 16 were excluded: seven papers were incomplete studies\nproviding no exact recurrence data (Abu-Ghazaleh et al , 1984;\nKilgore et al , 1984; Zuna et al , 1990; Morris et al , 1992; Patsner,\n1992; Trelford et al, 1995; Estape et al, 1998), one paper was a letter\nto the editor (Zuna, 1996), and another five studies lacked\nrelevance to prognosis and cervical cancer (Creasman and\nRutledge, 1971; Hughes et al , 1980; Ziselman et al , 1984; Willett,\n1985; Zuna and Behrens, 1996); three studies included patients\nwho underwent surgical treatment at advanced stages (Keettel et al,\n1974; Roberts et al , 1986; Imachi et al , 1987). The remaining six\nstudies reported on abnormal peritoneal cytology and prognosis\nand, thus, were included in the meta-analysis (Delgado et al, 1989;\nIto and Noda, 1992; Kashimura et al , 1997; Takeshima et al , 1997;\nKasamatsu et al , 2009; Kuji et al , 2014).\nStudy characteristics. After the final screening of the six relevant\nretrospective case–control observational studies, 1424 patients were\nenrolled in this meta-analysis, including a total of 64 cervical\ncancer patients with positive peritoneal cytology and 1360 with\nnegative cytology. Publication years of the eligible studies ranged\nfrom 1989 to 2014. The studies were from the United States and\nJapan, and were assessed to range from 7 to 8 on the nine-star\nNOS. Table 1 shows the study characteristics in detail.\nAll of the patients had FIGO stage IA-IIB disease. Most\nhistologic types, in particular, SCC, ADC and adenosquamous\nBRITISH JOURNAL OF CANCER Impact of positive peritoneal cytology on prognosis\n596 www.bjcancer.com | DOI:10.1038/bjc.2015.266\n\ncarcinoma, were included in each study, with the exception\nof two studies. Specifically, one study included only ADC histologic\ntype (Kasamatsu et al , 2009) and another, only SCC (Delgado\net al , 1989). Among all patients, 1360 (95.5%) showed negative\nperitoneal cytology, whereas positive peritoneal cytology was\nobserved in 64 (4.5%).\nMeta-analysis. Three studies (Ito and Noda, 1992; Takeshima\net al , 1997; Kasamatsu et al , 2009) included a total of 584 patients\nwith a combined total of 108 recurrences (20 of which occurred in\nthe 45 patients with positive peritoneal cytology). Figure 2A shows\nthe ORs for cervical cancer recurrence for each study, and for all\nstudies combined, comparing the positive peritoneal cytology with\nnegative cytology. As no heterogeneity existed among studies\n(P ¼ 0.473 and I2 ¼ 0%), the fixed-effects model was used. The\npooled OR for positive peritoneal cytology and the risk of cervical\ncancer recurrence was 4.468 (95% CI: 2.326–8.583, Po0.001).\nThere were six studies (Delgado et al, 1989; Ito and Noda, 1992;\nKashimura et al , 1997; Takeshima et al , 1997; Kasamatsu et al ,\n2009; Kuji et al , 2014), including 1424 patients with a combined\ntotal of 317 lymph node metastases (34 of which occurred in the 64\npatients with positive peritoneal cytology). Figure 2B shows the\nORs for lymph node metastasis for each study, and for all studies\ncombined, comparing positive peritoneal cytology with negative\ncytology. As no heterogeneity existed among studies ( P ¼ 0.606\nand I2 ¼ 0%), the fixed-effects model was used. The pooled OR for\npositive peritoneal cytology and the risk of lymph node metastasis\nwas 3.726 (95% CI: 2.127–6.525, Po0.001).\nIn the sensitivity analysis, the results based on the omission of\none study at a time and the calculation of the pooled OR for the\nremaining studies showed that no study had a significant effect on\nthe pooled OR (Figure 3A and B).\nSubgroup analysis by the type of histology. Figure 4A and B\nshows the OR for each study and the pooled OR for categories of\nADC (including adenosquamous carcinoma), SCC, and ADC\n(including adenosquamous carcinoma) plus SCC. Positive perito-\nneal cytology in ADC-type cervical cancer led to a significant\nincrease in recurrence (OR: 3.684; 95% CI: 1.680–8.077; P ¼ 0.001\nand I2 ¼ 0%). Positive peritoneal cytology in cervical cancer with\nthe ADC plus SCC histological type was also associated with\nrecurrence (OR: 9.667; 95% CI: 2.132–43.829; P ¼ 0.003 and\nI2 ¼ 0%). Although only one study (Takeshima et al , 1997)\nreported on the SCC histological type (OR: 1.701; 95% CI:\n0.172–16.780; P ¼ 0.649 and I2 ¼ 0%), there was no association\nbetween positive peritoneal cytology and recurrence (Figure 4A).\nPositive peritoneal cytology in ADC-type cervical cancer was\nassociated with a significant increase in lymph node metastasis\n(OR: 2.435; 95% CI: 1.088–5.450; P ¼ 0.030 and I2 ¼ 0%). Positive\nperitoneal cytology in cervical cancer with ADC plus SCC\nhistological type was also associated with lymph node metastasis\n(OR: 7.024; 95% CI: 2.422–20.368; Po0.001 and I2 ¼ 0%). As for\nthe SCC histological type, there were two studies (OR: 4.224; 95%\nCI: 1.343–13.288; P ¼ 0.014 and I2 ¼ 0%) (Figure 4B) (Delgado\net al , 1989; Kashimura et al , 1997).\nThese results suggest that positive peritoneal cytology is\nassociated with a higher risk of recurrence, especially for the\nADC or adenosquamous carcinoma histological types, and with\nhigher rates of lymph node metastasis, regardless of histological\ntype.\nPublication bias. A funnel plot for publication bias was slightly\nasymmetric (Figure 5A). This asymmetry could be caused by small\nstudy effects. The Begg and Mazumdar rank correlation test\nshowed no significance ( P ¼ 0.296). The Begg’s test presents some\nimportant limitations. A significant correlation suggests that a bias\nexists, but does not directly address the implications of this bias.\nConversely, a non-significant correlation may be due to low\nstatistical power, and cannot be considered as evidence that bias is\nabsent. In this study, the Egger’s test showed significance\n(P ¼ 0.038). The classic fail-safe N-method can address the\nconcerns that the entire observed effect may be an artefact of bias.\nThis meta-analysis incorporates data from three studies, which\nyielded a z-value of 4.64 and a corresponding two-tailed P-value of\n0.00001. The fail-safe N is 14. This means that we would need to\nlocate and include 14 ‘null’ studies in order for the combined\ntwo-tailed P-value to exceed 0.050. However, a funnel plot for\npublication bias regarding LN metastasis was almost symmetric\n(Figure 5B).\nArticles identified\nin Pubmed detabase search\n(n = 296)\nRecords afrer duplicates removed\n(n = 303)\nRecords screened\n(n = 22)\nFull-text articles assessed for eligibility\n(n = 22)\nIncluded in meta-analysis\n(n =6 )\nFull-text articles excluded\nLetter to the editor\nDid not give data to calculate\nNot relevant to prognosis\nNot relevant to cervical cancer\nIncluded advanced stage\n(n = 16)\n(n =1 )\n(n =7 )\n(n =1 )\n(n =4 )\n(n =3 )\nArticles identified\nin Cochrane detabase search\n(n =1 7 )\nArticles identified\nin EMBASE detabase search\n(n = 43)\nExcluded based on screening or titles\nand/or abstracts based on general criteria\n(n = 281)\nFigure 1. Flow diagram of the procedure for the literature search.\nImpact of positive peritoneal cytology on prognosis BRITISH JOURNAL OF CANCER\nwww.bjcancer.com | DOI:10.1038/bjc.2015.266 597\n\nTable 1. Characteristics of studies included in the meta-analysis\nStudy\nyear\nStudy\nperiod/country\nMean age\n(years)\nFIGO\nstage\nHistological\ntype\nMedian\nfollow-up\nperiod\n(months)\nPeritoneal\ncytology Sample size Recurrence LN metastasis\nAdjusted\nvariables\nKuji et al ,\n2014\nJapan/2002–2010 49.3 (20–75) 1B (179)\n2A (25)\n2B (24)\nSCC (139)\nADC (76)\nASC (13)\n51 (4–115) Positive Total (9) SCC (1)\nADC þ ASC (5)\nTotal (6)\nTotal (6) Peritoneal\ncytology,\nhistological\ntype, LN\nmetastasis, LVSI,\nparametrium\ninvasion, deep\nstromal invasion,\nuterine body\ninvasion, ovarian\nmetastasis\nNegative Total (219) ADC þ ASC (11) Total (51)\nKasamatsu\net al , 2009\nJapan/1984–2003 48 1B–2B ADC (107) 72 (1–281) Positive ADC (16) ADC (8) ADC (5) Peritoneal\ncytology, LN\nmetastasis, LVSI,\ntumour size,\ndepth in cervical\nwall,\nparametrium\ninvasion,\ninfiltration to\nvaginal, ovarian\nmetastasis,\nhistlogical grade\nNegative ADC (91) ADC (16) ADC (17)\nTakeshima\net al , 1997\nJapan/1982–1993 NM 1B–2B SCC\nADC\nASC\n(36–168) Positive SCC (4)\nADC þ ASC (15)\nTotal (19)\nSCC (1)\nADC þ ASC (7)\nTotal (8)\nADC þ ASC (7) Age, peritoneal\ncytology, LN\nmetastasis,\nvessel\npermeation,\nmuscle invasion,\novarian\nmetastasis,\nparametrial\ninvasion, stage\nNegative SCC (238)\nADC þ ASC (117)\nTotal (355)\nSCC (39)\nADC þ ASC (27)\nTotal (66)\nADC þ ASC (27)\nKashimura\net al , 1997\nJapan/1978–1994 NM 1B–2B SCC\nADC\nNM Positive SCC (12)\nADC (10)\nTotal (22)\nNR SCC (8)\nADC (7)\nTotal (15)\nAge, histology,\nLN metastasis,\npara-aortic LN\nmetastasis,\novarian\nmetastasis,\nperitoneal\ncytology\nNegative SCC (215)\nADC (49)\nTotal (264)\nNR SCC (59)\nADC (13)\nTotal (72)\nIto and\nNoda, 1992\nJapan/1978–1994 (34–77) 2 SCC\nADC\nASC\n(0–70) Positive Total (10) Total (4) Total (8)\nNegative Total (93) Total (6) Total (32)\nDelgado\net al , 1989\nUS/1981–1984 NM 1 SCC NM Positive SCC (2) NR SCC (0) Age, LVSI, depth\nof invasion,\nparametrial\ninvasion\nNegative SCC (625) NR SCC (97)\nAbbreviations: ADC ¼ adenocarcinoma; ASC ¼ adenosquamous carcinoma; FIGO ¼ International Federation of Gynecology and Obstetrics; LN ¼ lymph node; LVSI ¼ lymphovascular space\ninvasion; NM ¼ not mentioned; NR ¼ not reported; SCC ¼ squamous cell carcinoma.\nBRITISH JOURNAL OF CANCER Impact of positive peritoneal cytology on prognosis\n598 www.bjcancer.com | DOI:10.1038/bjc.2015.266\n\nStatistics for each study\nStatistics for each study\nStudy name\nStudy name\nIto, 1992\nIto, 1992\nTakeshima, 1997\nTakeshima, 1997\nKasamatsu, 2009\nKasamatsu, 2009\nKuji, 2014\nDelgado, 1989\nKashimura, 1997\nZ-value P- value\nP- value\nOdds\nratio\nOdds\nratio\nLower\nlimit\nLower\nlimit\nUpper\nlimit\nUpper\nlimit\nOdds ratio and 95% Cl\nOdds ratio and 95% Cl\nCytology(–)\nCytology(–)\n0.1\n0.01\n0.0006.525\n27.280\n6.447\n8.778\n18.220\n38.050\n22.7550.0521.084\n7.625\n5.288\n2.917\n1.979\n6.588\n3.726\n1.528\n1.535\n0.969\n0.607\n1.591\n2.127\n0.009\n0.258\n0.057\n0.008\n0.013\n0.959\n0.1\n9.667 2.132 2.942 0.003\n0.017\n0.007\n0.000\n2.392\n2.706\n4.495\n43.829\n8.227\n14.349\n8.583\n1.233\n1.531\n2.326\n3.185\n4.688\n4.468\n0.2 0.5 1\n1 10 100\n25 10\nCytology(+)\nCytology(+)\nA\nB\nFigure 2. The odds ratio for the risk of ( A) recurrence and ( B) LN metastasis for each study and all studies combined comparing positive\nperitoneal cytology with negative cytology in a meta-analysis based on the fixed-effects model. Heterogeneity was low across studies\n(A: P ¼ 0.473 and I2 ¼ 0%; B: P ¼ 0.606 and I2 ¼ 0%). The size of each square is proportional to the sample size for each study, and the horizontal\nline through the square indicates the 95% confidence interval for that study. For the pooled analysis, the diamond indicates the pooled value, and\nthe right and left ends of the diamond indicate the 95% confidence interval for the analysis.\nStatistics with study removed\nStatistics with study removed\nStudy name\nStudy name\nIto, 1992\nIto, 1992\nTakeshima, 1997\nTakeshima, 1997\nKasamatsu, 2009\nKasamatsu, 2009\nKuji, 2014\nDelgado, 1989\nKashimura, 1997\nZ-value P- value\nP- value\nPoint\nPoint\nLower\nlimit\nLower\nlimit\nUpper\nlimit\nUpper\nlimit\nOdds ratio (95% Cl)\nwith study removed\nOdds ratio (95% Cl)\nwith study removed\nCytology (–)\nCytology (–)\n0.1\n0.01\n0.000\n0.000\n0.000\n0.000\n0.000\n0.000\n0.000\n6.525\n6.172\n8.462\n7.780\n6.382\n6.135\n6.8812.200\n3.374\n3.891\n3.403\n4.058\n4.477\n3.354\n3.726\n1.856\n1.815\n2.117\n2.369\n1.823\n2.127\n0.1\n3.744 1.815 3.575\n0.000\n0.000\n0.000\n0.000\n3.925\n3.590\n4.495\n7.721\n14.885\n9.739\n8.583\n2.464\n1.951\n2.326\n6.056\n4.359\n4.468\n0.01 1\n1 10 100\n10010\nCytology (+)\nCytology (+)\nA\nB\nFigure 3. Sensitivity analysis of the summary odds ratio coefficients on the relationships between abnormal peritoneal cytology and prognosis\n(A: recurrences, B: LN metastasis) of patients with cervical cancer.\nImpact of positive peritoneal cytology on prognosis BRITISH JOURNAL OF CANCER\nwww.bjcancer.com | DOI:10.1038/bjc.2015.266 599\n\nDISCUSSION\nAccording to published data, patients with positive peritoneal\ncytology had poorer prognostic outcomes, such as survival rates\nand recurrence, than those with negative cytology. However, the\nsample size in most studies was not large enough to sufficiently\nexplain the effectiveness of cytology on cervical cancer prognosis,\nand no quantitative analysis has estimated the efficacy of cytology\non cervical cancer prognosis. Moreover, it is well known that,\npositive peritoneal cytology overlaps with other risk factors.\nRoberts et al (1986) reported that positive peritoneal cytology\nwas associated with a very poor prognosis, and it was also\nassociated with other poor prognostic factors (Roberts et al, 1986).\nThus, no consensus was reached as to whether positive peritoneal\ncytology is an independent risk factor. Takeshima et al (1997)\nfound that peritoneal cytology was not a prognostic factor for\nstage I and II ADC, although cardinal ligament invasion, lymph\nnode metastases, and muscle layer invasion were prognostic factors\n(Takeshima et al , 1997). Kasamatsu et al (2009) found that\nperitoneal cytology, para-aortic lymph nodes, and pelvic lymph\nnodes were independent prognostic factors, irrespective of the\nhistological type (Kasamatsu et al, 2009). Kuji et al (2014) reported\nthat positive peritoneal cytology may be associated with a poor\nprognosis in adenosquamous carcinoma or ADC of cervix (Kuji\net al , 2014). The results of the present meta-analysis indicate that\npositive peritoneal cytology of cervical cancer patients leads to a\nsignificant increase in the risk of cervical cancer recurrence.\nHowever, the results of the subgroup analysis based on the\nhistology show slightly different patterns. The estimates from the\nADC histology type indicate that positive peritoneal cytology leads\nto an increase in the risk of recurrence (OR: 3.684; 95% CI: 1.680–\n8.077; P ¼ 0.001), whereas those from the squamous cell type did\nnot lead to a significant increase in the risk of recurrence (OR:\n1.701; 95% CI: 0.172–16.780; P ¼ 0.649). This finding might\nsuggest the possibility that positive peritoneal cytology increases\ndisease recurrence by the spread of viable tumour cells. If we can\nconclude that positive peritoneal cytology, particularly for patients\nwith ADC, leads to recurrence by peritoneal dissemination through\nhematogenous or lymphatic spread, then positive peritoneal\ncytology would indicate systemic disease. According to the\nqualitative analysis, positive results of peritoneal cytology seem\nto be associated with high recurrence rates through peritoneal\ndissemination. The authors reported that the incidence of\nperitoneal spread at the first recurrence in the positive cytology\ngroup (62.5%) was significantly higher than that in the negative\ncytology group (12.5%) (Kasamatsu et al , 2009). In addition,\nanother study reported that peritoneal recurrence of ADC among\npatients with positive peritoneal cytology occurred in 60% of cases\n(Kuji et al, 2014). This percentage tended to be higher than that in\npatients with negative cytology. However, in the present meta-\nanalysis, we could not conduct a quantitative meta-analysis on the\nGroup by\nhistology\nGroup by\nhistology\nStudy name\nStudy name\nSubgroup within study\nStatistics for each study\nStatistics for each study Odds ratio and 95% Cl\nOdds ratio and 95% Cl\nAdeno\nAdeno Takeshima, 1997\nKasamatsu, 2009\nIto, 1992\nKuji, 2014\nDelgado, 1989\nKashimura, 1997\nAdeno\nAdeno\nBoth\nBoth\nBoth\nSquamous\nSquamous\nSquamous\nAdeno\nAdeno\nBoth\nBoth\nSquamous\nSquamous\nTakeshima, 1997\nTakeshima, 1997\nKasamatsu, 2009\nIto, 1992 Both\n2.917\nOdds\nratio\n4.688\n3.684\n9.667\n9.667\n1.701\n1.701\n0.01 0.1 11 0 100\nOdds\nratio\n2.917\n1.979\n2.435\n7.625\n6.588\n7.024\n1.084\n5.288\n4.224\nLower\nlimit\n1.343\n0.052\n1.535\n2.422\n1.591\n1.528\n1.088\n0.607\n0.969\nUpper\nlimit\n8.778\n6.447\n5.450\n38.050\n27.280\n20.368\n22.755\n18.220\n13.288\nP-Value\n0.014\n0.008\n0.000\n0.959\n0.009\n0.013\n0.030\n0.258\n0.057\nLower\nlimit\n0.172\n0.172\n2.132\n2.132\n1.680\n1.531\n0.969\nP-Value\n0.057\n0.007\n0.001\n0.003\n0.003\n0.649\n0.649\n0.01 0.1 1 10 100\nCytology (+)Cytology (–)\nCytology (+)Cytology (–)\nUpper\nlimit\n8.778\n14.349\n8.077\n43.829\n43.829\n16.780\n16.780\nAdeno\nAdeno\nSquamous\nB\nA\nFigure 4. Positive peritoneal cytology and the risk of ( A) recurrence and ( B) LN metastasis in the subgroup meta-analysis by histological type,\nbased on the fixed-effects model.\nBRITISH JOURNAL OF CANCER Impact of positive peritoneal cytology on prognosis\n600 www.bjcancer.com | DOI:10.1038/bjc.2015.266\n\nrelationship between abnormal peritoneal cytology and metastatic\npattern including peritoneal dissemination, because we could not\nobtain detailed individual data. Considering the above facts, further\ninvestigation and well-designed randomised controlled trials may\nbe needed to reach a definitive conclusion.\nDelgado et al (1989) found that the factors identified as\nindependent risk factors for pelvic lymph node metastasis by\nmultivariate analysis were depth of invasion, parametrium\ninvolvement, lymph-vascular space involvement, and age, but not\nperitoneal cytology (Delgado et al , 1989). Moreover, Ito and Noda\n(1992) reported that negative cytology was often found even in\ncases with metastasis to retroperitoneal lymph nodes, whereas\npositive cytology was found in cases without metastasis (Ito and\nNoda, 1992). These authors suggested that the mechanism of\ncervical carcinoma metastasis to retroperitoneal lymph nodes\nmight not be the same as that of spreading into the abdominal\ncavity. However, Kashimura et al (1997) reported that peritoneal\ncytology, lymph node metastasis, ovarian metastasis, and histolo-\ngical grade are independent prognostic factors in stage I and\nstage II ADC of the cervix (Kashimura et al , 1997). The results\nof our meta-analysis indicate that positive peritoneal cytology of\ncervical cancer patients leads to a significant increase in the risk of\nlymph node metastasis. This pattern was also observed in the\nsubgroup analysis for histological type. The estimates from the\nADC histology type and SCC type indicate that positive peritoneal\ncytology leads to an increase in the risk of lymph node metastasis\n(OR: 2.435; 95% CI: 1.088–5.450; P ¼ 0.030, and OR: 4.224; 95%\nCI: 1.343–13.288; P ¼ 0.014, respectively). This finding could\npresent, whatever histology type of cancer was, significant increase\nin lymph node metastasis was found between the peritoneal\ncytology positive and cytology negative groups.\nOur meta-analysis results revealed that positive peritoneal\ncytology was associated with recurrence in histology type including\nADC or adenosquamous carcinoma and lymph node metastasis\nregardless of histological type. Therefore, it is crucial to obtain\nperitoneal cytology fluid routinely at the time of operation in patients\nwith early-stage cervical cancer, especially ADC histological type.\nAlthough we performed adjuvant treatment such as postoperative\nconcurrent chemoradiotherapy or radiotherapy according to the risk\nfactors, if positive peritoneal cytology truly reflects peritoneal spread\nor affects survival adversely, we think that systemic chemotherapy\ncould be considered as the treatment of choice for patients with\npositive peritoneal cytology, especially for those with the ADC\nhistological type, for similar reasons regarding peritoneal cytology as\na risk factor of endometrial and ovarian cancers (Rahaman et al ,\n2009; Berek, 2012a, b). Patients with positive cytology along with\npoor prognostic factors may potentially benefit from systemic\nchemotherapy in endometrial cancer (Berek, 2012b).\nNevertheless, several limitations of the present meta-analysis\nshould be recognised. First, as our meta-analysis was a retro-\nspective study, it may have led to subject selection bias, which\ncould have influenced the reliability of our results. Second, this\nmeta-analysis was based on data published from the included\nstudies, and we could not obtain detailed individual data. Thus, we\nwere not able to correct all clinicopathological parameters into a\nconsistent standard, for example, stage, risk factors (lymph node\nstatus, parametrial invasion, operative margin status, tumour size,\nlymphovascular space invasion, and deep stromal invasion),\ntreatment method (e.g., radical operation field scale or adjuvant\ntreatment method), follow-up period, recurrence diagnostic\nmethod, and recurrent pattern. Therefore, it was not possible to\nperform multivariate analysis and to determine the independent\nassociated risk factor of poor prognosis among patients with\npositive peritoneal cytology and lymph node involvement in the\npresent meta-analysis. In addition, we could not control potentially\nrelevant bias factors (i.e., age, and ethnic group). Finally, funnel\nplot analyses concerning publication bias were also performed to\nconfirm the reliability of our research results. In our meta-analysis,\nthe publication bias in positive peritoneal cytology with recurrence\nmay exist, but not with lymph node metastasis in cervical cancer.\nConsequently, more studies should be performed so that they can\nbe included in a meta-analysis.\nAlthough our study has some limitations, we believed it is the\nfirst meta-analysis focusing on the relationships between peritoneal\ncytology and the prognosis of cervical cancer. Furthermore, we\nmade an effort to perform a highly sensitive literature search\nstrategy utilising electronic databases.\nIn summary, to our knowledge, our meta-analysis is the first\nreport to show that positive peritoneal cytology may be strongly\nassociated with poor prognosis in cervical cancer. Our meta-analysis\nindicates that peritoneal cytology may serve as a prognostic factor for\ncervical cancer. Thus, it is essential to obtain peritoneal cytology\nroutinely at the time of laparotomy in patients with cervical cancer\nabout histology type including ADC or adenosquamous carcinoma,\nespecially. Moreover, the results of peritoneal cytology must be\nconsidered in postoperative treatment planning. Therefore, we\npropose that positive cytology should be reported separately without\nchanging the stage, and suggest that positive peritoneal cytology is a\nrisk factor that should be taken into account when making decisions\nconcerning postoperative adjuvant therapy in patients with early-\nstage cervical cancer. When the result of peritoneal cytology is\npositive, additional aggressive postoperative chemotherapy may be\nproposed as an adjuvant therapy.\nHowever, owing to the limitations acknowledged above, further\nresearch, with larger sample sizes and more amount of compre-\nhensive data, is still required to provide a more representative\nstatistical analysis. Therefore, well-designed randomised controlled\ntrials or prospective cohort studies are needed in the future.\n0.0\n0.0\n0.5\n1.0\n1.5\n2.0\n0.2\n0.4s.e.s.e.\n0.6\n0\nLog odds ratio\nLog odds ratio\n123\n0.8\n–3 –2 –1\n0123–3 –2 –1\nA\nB\nFigure 5. Assessment of publication bias using funnel plot analysis.\n(A) Funnel plot analysis of studies on recurrence. The funnel plot has a\nasymmetrical distribution, suggesting the presence of selection bias.\n(B) Funnel plot analysis of studies on LN metastasis. The Funnel plot has\na almost symmetrical distribution. Publication bias was not found in the\nmeta-analysis of LN metastasis.\nImpact of positive peritoneal cytology on prognosis BRITISH JOURNAL OF CANCER\nwww.bjcancer.com | DOI:10.1038/bjc.2015.266 601\n\nACKNOWLEDGEMENTS\nThis work was supported by the Konkuk University. This study\nreceived no financial support.\nCONFLICT OF INTEREST\nThe authors declare no conflict of interest.\nREFERENCES\nAbu-Ghazaleh S, Johnston W, Creasman WT (1984) The significance of\nperitoneal cytology in patients with carcinoma of the cervix. Gynecol Oncol\n17(2): 139–148.\nBegg CB, Mazumdar M (1994) Operating characteristics of a rank correlation\ntest for publication bias. Biometrics 50(4): 1088–1101.\nBenedet JL, Odicino F, Maisonneuve P, Beller U, Creasman WT, Heintz AP,\nNgan HY, Pecorelli S (2003) Carcinoma of the cervix uteri. Int J Gynaecol\nObstet 83(Suppl 1): 41–78.\nBerek JS (2012a) Berek & Novak’s Gynecology , 15th edn, vol. ch. 37.\nLippincott Williams & Wilkins: Philadelphia, PA, USA.\nBerek JS (2012b) Berek & Novak’s Gynecology , 15th edn. Lippincott Williams\n& Wilkins: Philadelphia, PA, USA.\nCreasman WT, Rutledge F (1971) The prognostic value of peritoneal cytology\nin gynecologic malignant disease. Am J Obstet Gynecol 110(6): 773–781.\nDelgado G, Bundy BN, Fowler Jr WC, Stehman FB, Sevin B, Creasman WT,\nMajor F, DiSaia P, Zaino R (1989) A prospective surgical pathological\nstudy of stage I squamous carcinoma of the cervix: a Gynecologic\nOncology Group Study. Gynecol Oncol 35(3): 314–320.\nEstape R, Angioli R, Wagman F, Madrigal M, Janicek M, Ganjei-Azar P,\nPenalver M, Averette H (1998) Significance of intraperitoneal cytology in\npatients undergoing radical hysterectomy. Gynecol Oncol 68(2): 169–171.\nHiggins JP, Thompson SG (2002) Quantifying heterogeneity in a meta-\nanalysis. Stat Med 21(11): 1539–1558.\nHiggins JP, Thompson SG, Deeks JJ, Altman DG (2003) Measuring\ninconsistency in meta-analyses. BMJ 327(7414): 557–560.\nHughes RR, Brewington KC, Hanjani P, Photopulos G, Dick D, Votava C,\nMoran M, Coleman S (1980) Extended field irradiation for cervical cancer\nbased on surgical staging. Gynecol Oncol 9(2): 153–161.\nImachi M, Tsukamoto N, Matsuyama T, Nakano H (1987) Peritoneal cytology\nin patients with carcinoma of the uterine cervix. Gynecol Oncol 26(2):\n202–207.\nIto K, Noda K (1992) Peritoneal cytology in patients with uterine cervical\ncarcinoma. Gynecol Oncol 47(1): 76–79.\nKasamatsu T, Onda T, Sasajima Y, Kato T, Ikeda S, Ishikawa M, Tsuda H\n(2009) Prognostic significance of positive peritoneal cytology in\nadenocarcinoma of the uterine cervix. Gynecol Oncol 115(3): 488–492.\nKashimura M, Sugihara K, Toki N, Matsuura Y, Kawagoe T, Kamura T,\nKaku T, Tsuruchi N, Nakashima H, Sakai H (1997) The significance of\nperitoneal cytology in uterine cervix and endometrial cancer. Gynecol\nOncol 67(3): 285–290.\nKeettel WC, Pixley EE, Buchsbaum HJ (1974) Experience with peritoneal\ncytology in the management of gynecologic malignancies. Am J Obstet\nGynecol 120(2): 174–182.\nKilgore LC, Orr Jr JW, Hatch KD, Shingleton HM, Roberson J (1984)\nPeritoneal cytology in patients with squamous cell carcinoma of the cervix.\nGynecol Oncol 19(1): 24–29.\nKuji S, Hirashima Y, Komeda S, Tanaka A, Abe M, Takahashi N,\nTakekuma M (2014) The relationship between positive peritoneal cytology\nand the prognosis of patients with FIGO stage I/II uterine cervical cancer.\nJ Gynecol Oncol 25(2): 90–96.\nMorris PC, Haugen J, Anderson B, Buller R(1992) The significance of peritoneal\ncytology in stage IB cervical cancer. Obstet Gynecol 80(2): 196–198.\nOrwin RG, Boruch RF (1983) RRT meets RDD: statistical strategies for assuring\nresponse privacy in telephone surveys. Public Opin Q 46(4): 560–571.\nPatsner B (1992) Peritoneal cytology in patients with stage IB cervical cancer\nundergoing radical hysterectomy: limited value. Eur J Gynaecol Oncol\n13(4): 306–308.\nRahaman J, Steiner N, Hayes MP, Chuang L, Fishman D, Gretz Iii H (2009)\nChemotherapy for Gynecologic cancers. Mt Sinal J Med 76: 577–588.\nRoberts WS, Bryson SC, Cavanagh D, Roberts VC, Lyman GH (1986)\nPeritoneal cytology and invasive carcinoma of the cervix. Gynecol Oncol\n24(3): 331–336.\nStang A (2010) Critical evaluation of the Newcastle-Ottawa scale for the\nassessment of the quality of nonrandomized studies in meta-analyses.\nEur J Epidemiol 25(9): 603–605.\nSterne JA, Egger M (2001) Funnel plots for detecting bias in meta-analysis:\nguidelines on choice of axis. J Clin Epidemiol 54(10): 1046–1055.\nSterne JA, Egger M, Smith GD (2001) Systematic reviews in health care:\nInvestigating and dealing with publication and other biases in meta-\nanalysis. BMJ 323(7304): 101–105.\nStroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D,\nMoher D, Becker BJ, Sipe TA, Thacker SB (2000) Meta-analysis of\nobservational studies in epidemiology: a proposal for reporting.\nMeta-analysis Of Observational Studies in Epidemiology (MOOSE) group.\nJAMA 283(15): 2008–2012.\nTakeshima N, Katase K, Hirai Y, Yamawaki T, Yamauchi K, Hasumi K (1997)\nPrognostic value of peritoneal cytology in patients with carcinoma of the\nuterine cervix. Gynecol Oncol 64(1): 136–140.\nTrelford JD, Kinney W, Vogt P (1995) Positive peritoneal cytology in stage I\ncarcinoma of the cervix. Acta Cytol 39(2): 177–179.\nWillett GD (1985) Prognostic value of cytologic peritoneal washings. Clin Lab\nMed 5(2): 265–274.\nZiselman EM, Harkavy SE, Hogan M, West W, Atkinson B (1984) Peritoneal\nwashing cytology. Uses and diagnostic criteria in gynecologic neoplasms.\nActa Cytol 28(2): 105–110.\nZuna RE (1996) Peritoneal washing cytology in gynecologic cancer: long-term\nfollow-up of 355 patients [Letter to the editor]. J Natl Cancer Inst 88(23):\n1779.\nZuna RE, Behrens A (1996) Peritoneal washing cytology in gynecologic\ncancers: long-term follow-up of 355 patients. J Natl Cancer Inst 88(14):\n980–987.\nZuna RE, Hansen K, Mann W (1990) Peritoneal washing cytology in cervical\ncarcinoma. Analysis of 109 patients. Acta Cytol 34(5): 645–651.\nZuna RE, Moore WE, Shanesmith RP, Dunn ST, Wang SS, Schiffman M,\nBlakey GL, Teel T (2009) Association of HPV16 E6 variants with\ndiagnostic severity in cervical cytology samples of 354 women in a US\npopulation. Int J Cancer 125(11): 2609–2613.\nThis work is published under the standard license to publish agree-\nment. After 12 months the work will become freely available and\nthe license terms will switch to a Creative Commons Attribution-\nNonCommercial-Share Alike 4.0 Unported License.\nBRITISH JOURNAL OF CANCER Impact of positive peritoneal cytology on prognosis\n602 www.bjcancer.com | DOI:10.1038/bjc.2015.266","source_license":"CC0","license_restricted":false}