Case
A 45-year-old nulligravid woman with a medical history significant for uterine fibroids, ovarian cysts, and dysmenorrhea presented to the emergency department with severe right lower quadrant pain radiating to her right leg, two weeks after receiving PRP injections into the ovaries by an external provider. The procedure was intended to achieve “ovarian rejuvenation” and improve fertility potential.
Her symptoms began approximately one week before presentation and included chills, decreased appetite, and progressively worsening pelvic pain. She was started empirically on oral amoxicillin-clavulanate and doxycycline by her outside provider, without significant improvement.
On presentation to the hospital, the patient was afebrile and hemodynamically stable. Physical examination revealed diffuse lower abdominal and cervical motion tenderness, and bilateral adnexal tenderness. Laboratory findings were notable for leukocytosis (white blood cell count 18.1 × 10 3 /μL) and mild lactic acidosis (lactate 2.1 mmol/L). Computed tomography demonstrated multiple right ovarian cystic structures, measuring 6.1 cm and 3.1 cm, as well as a left ovarian cyst, measuring up to 3.9 cm in greatest dimension ( Fig. 1 A). Similarly, transvaginal ultrasound showed multiple heterogeneous cystic structures in both ovaries, measuring approximately 5.7 cm and 5.4 cm on the right, as well as 3.8 cm on the left, with preserved Doppler flow bilaterally ( Fig. 1 B). Figure 1 Imaging Findings. Computed tomography (A) and transvaginal ultrasound (B) images of the patient who presented with tubo-ovarian abscesses after PRP injections.
Imaging Findings. Computed tomography (A) and transvaginal ultrasound (B) images of the patient who presented with tubo-ovarian abscesses after PRP injections.
She was admitted for presumed pelvic inflammatory disease with concern for tubo-ovarian abscess vs. superinfected endometriomas. Empiric intravenous antibiotics (ceftriaxone, doxycycline, and metronidazole) were initiated. Her clinical picture improved; she was afebrile for 72 hours, leukocytosis and lactic acidosis resolved, and she was discharged on hospital day 3 with a 14-day course of doxycycline and metronidazole.
At her scheduled outpatient follow-up two weeks later, she reported persistent debilitating pelvic pain. She described sharp lower quadrant pain with movement, exacerbated by routine activities, such as riding in a car or using elevators. Given these findings, she was advised to return to the emergency department. On her second presentation, she remained afebrile with a normal white blood cell count (8.1 x 10 3 /μL), but pelvic examination again revealed cervical motion and bilateral adnexal tenderness. Repeat imaging showed unchanged multicystic ovarian structures. She was restarted on IV antibiotics. On day 2 of admission, a fever of 38.2°C developed. Given the lack of clinical improvement, interventional radiology (IR) was consulted. The IR team performed image-guided transabdominal aspiration of the bilateral cystic structures using an 18-gauge needle, yielding 20 mL of purulent fluid that led to near-complete drainage of the collections without the need for drain placement. Culture results confirmed Streptococcus anginosus . The patient remained afebrile for 48 hours after the procedure and was discharged with a 14-day course of oral doxycycline and metronidazole. At follow-up four months later, she reported complete resolution of pelvic pain, return to regular menses, and resumption of her normal activities and work. Written informed consent was obtained from the patient for publication of this case report and any accompanying images.
Credit
Vanessa Kalinowska: Writing – original draft, Methodology, Investigation, Data curation, Conceptualization. Haaris Mobin: Writing – original draft, Investigation. Elnur Babayev: Writing – review & editing, Validation, Supervision, Investigation.
Discussion
This case illustrates a serious complication following intraovarian PRP injection, a procedure that is offered for fertility enhancement in some practices despite limited clinical validation. Although PRP is theorized to stimulate follicular activation and angiogenesis, the evidence on its efficacy is limited and recent randomized control trials shows no benefit ( 13 ). Moreover, ovarian injections pose an infection risk. Transvaginal ovarian access has inherent infection risks, albeit very small. The risk of infection for a transvaginal egg retrieval is reported at <0.5% ( 15 ). In addition to this risk, PRP injections involve the ovarian introduction of a blood product, which likely increases the infection risk. Strept. anginosus , isolated from the aspirated abscesses in our case, is known for its association with abscess formation and may represent vaginal or skin flora introduced during the PRP injection ( 16 ).
Restricted access to prior medical records from external institutions, including previous imaging studies, specifics of the ovarian PRP injection protocol, and whether this injection was performed as part of the clinical trial enrollment, limited our ability to describe the patient’s earlier care comprehensively. However, the patient reported a known history of ovarian cysts and denied being involved in a clinical study. Moreover, our review of the external institution’s website where the patient received these injections indicated that this treatment is offered as a clinical service, with no mention of any association with a clinical trial. This institution also is not listed on ClinicalTrials.gov as a participating site for such trials. Taken together, these observations suggest that the ovarian PRP injections were not administered as part of a registered clinical trial.
Post-procedural imaging revealed complex, heterogenous, bilateral ovarian cystic structures. These were aspirated by the IR team in our case. However, depending on institutional resources and the reproductive endocrinology and infertility specialist’s training and procedural expertise, tubo-ovarian abscesses also may be drained by reproductive endocrinology and infertility physicians, particularly when a transvaginal approach is feasible. In this case, it also is unclear if these abscesses arose de novo as infectious masses or represented superinfected pre-existing endometriomas. Given the patient's history of dysmenorrhea, underlying endometriosis is a reasonable suspicion. If endometriomas were present before the procedure, this patient should have been cautioned against proceeding, given a recent case study illustrating the risk of pelvic inflammatory disease and subsequent bacteremia in a patient with endometriosis undergoing intraovarian PRP injections ( 7 ).
To date, most published reports on ovarian PRP emphasize potential benefits, often without rigorous follow-up or safety data. Adverse events remain underreported, and there are no standardized protocols for preparation, administration, or patient selection ( 17 ). This case reinforces the need for prospective clinical trials, standardized guidelines, and regulatory oversight before intraovarian PRP can be considered a safe and evidence-based option in reproductive medicine.
In conclusion, PRP injection to the ovaries is a relatively novel but unvalidated intervention for fertility enhancement that may carry significant risks. Clinicians should maintain a high index of suspicion for infection in patients presenting with pelvic pain after such procedures, and patients should be fully informed of potential complications. Further research is essential to determine the efficacy and safety of ovarian PRP. Given the risks of this procedure, intraovarian PRP injections to improve infertility treatment outcomes can be administered only under the careful guidance and regulation of a clinical trial until further research confirms its safety and efficacy. Adverse events should be reported thoroughly in these trials.
Coi Statement
V.K. has nothing to disclose. H.M. has nothing to disclose. E.B. has nothing to disclose.
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.