NAERS has been a valuable asset in monitoring and evaluating the quality and safety of the PEPFAR VMMC program. Information gathered has contributed to changes and improvements in VMMC programs and has informed WHO guidance. Examples of these lessons learned are included in Table 2. With data combined from many countries, rare but serious NAEs that otherwise may be unnoticed can be detected and quantified. As part of the follow-up for each reported case, PEPFAR makes recommendations regarding program improvement (country-specific or across countries, as relevant) so that future NAEs can be reduced. Additional actions for improvement may also occur within specific countries through a MOH review of these same AEs or local changes at the implementing partner, facility, or individual provider levels.

The early reports of deaths, whether determined related to the VMMC procedure or not, prompted standardization of an emergency resuscitation equipment list and the requirement for all facilities to have immediate access to these potentially life-saving medications and equipment. In-depth quality assurance tools were developed to assess compliance with this requirement along with adherence to multiple other quality standards.

Infection is the most common NAE type reported in the system. Recognition of this has prompted increased attention to infection prevention and control activities in PEPFAR guidance and technical assistance, including the development of a surgical site infection prevention, diagnosis, and management guide for VMMC programs. Information gathered from NAERS, particularly on necrotizing wound infections led to the detailed discussion of the diagnosis and management of Fournier’s Gangrene in the adverse event action guide [16]. There is now emphasis on the need for early detection and aggressive treatment with both broad-spectrum antibiotics and emergent referral to a higher level of surgical care for debridement. Reported cases of infection following episodes of post-operative bleeding demonstrate the risk of infection posed by blood sequestered in tissue. Investigation and management of these cases led to guidance in the action guide recommending increased vigilance for signs and symptoms of infection in those with post-operative hematomas [16].

The reported cases of fatal and non-fatal tetanus following VMMC demonstrated the vulnerability of young males in countries that do not include tetanus booster doses to males in immunization schedules, including most of the 15 countries prioritized for VMMC [17, 18]. Reports of the use of traditional remedies on surgical wounds prior to the onset of tetanus led to enhanced counseling to avoid applying any substances to the wound and an emphasis on the clean care approach. Data shared with WHO allowed analysis demonstrating an increased risk of tetanus with the use of an elastic collar compression VMMC device compared to surgery. This led WHO to recommend the use of elastic collar compression devices only with assurance of protective immunity against tetanus through vaccination including administration of two doses, at least 4 weeks apart with the last dose at least 2 weeks prior to the procedure, in clients without known previous tetanus immunization [19]. Using data on risk, WHO and PEPFAR encouraged national immunization programs to add tetanus booster doses for males to their immunization schedules and encouraged the use of VMMC platforms for the administration of these boosters; PEPFAR also indicated support for MOH implementation of tetanus mitigation and treatment strategies for VMMC clients [19].

Bleeding NAE cases highlight the importance of screening for potential bleeding disorders [20]. Because many clients may have never had a prior medical procedure, VMMC may be the first opportunity to recognize an undiagnosed bleeding dyscrasia. Following cases with severe bleeding likely due to a coagulopathy, and lack of recognition of such at the time of postoperative bleeding, screening materials were modified to include additional questions on individual and family histories of bleeding [21]. In these materials, providers were advised to re-question clients/parents upon presentation with any postoperative bleeding since a previously unrevealed bleeding disorder may be noted at the time of the bleeding. Due to the familial X-linked nature of Hemophilia A, one of the most common bleeding abnormalities, providers were advised not to perform VMMC on the brothers and cousins related through maternal aunts of clients with a bleeding disorder without proper preoperative evaluation.

In some cases, clients with significant bleeding were not referred for hospital admission until they presented several times with recurrent bleeding. Several clients demonstrated significant blood loss by the time of referral, demonstrating a need for more timely intervention. To ensure appropriate and timely referral of clients presenting with more than one episode of bleeding, algorithms for the management of postoperative bleeding were developed and included in the Action Guide [16]. Because clients with bleeding presented to VMMC clinics as well as clinics not staffed by providers trained in VMMC, separate algorithms with criteria for management and referral of clients with postoperative bleeding were developed for both VMMC clinics and other health care settings.

Initial observations of severe glans injuries and amputations in young clients following the use of the forceps-guided technique demonstrated the risk of injury with this procedure in young and sexually immature clients in whom there cannot be reliable palpation of the glans through the foreskin prior to foreskin excision. This led to WHO guidance and PEPFAR policy change to mandate the replacement of the widely used forceps-guided technique with the dorsal slit technique in clients under the age of 15 years. While this required widespread provider retraining, the seriousness of the glans injuries required this change to assure safety. When occasional cases continued to be reported with the incorrect use of the forceps-guided method, PEPFAR followed up with country teams and partners to ensure that providers were properly trained/re-trained in the use of dorsal slit surgical method and modified surgical instrument packs to include only the appropriate instruments for dorsal slit procedures while eliminating instruments necessary to perform the forceps-guided technique. Multiple glans injury cases submitted to NAERS occurred during early infant male circumcision (EIMC). The only method used, the Mogen clamp, was like the forceps-guided technique in that it did not allow glans visualization after clamping and before foreskin excision.

Analysis of fistula cases reported in NAERS also demonstrated an increased risk in young clients with immature genitalia [22]. Additional clinical information collected during NAERS reporting provided plausible hypotheses for mechanisms of urethral injury during circumcision, with age, likely as a proxy for physical maturity, as the most concerning. Although both glans injuries and fistulas were very rare occurrences, the increased risk in young clients led PEPFAR to change policy starting October 2020 to stop support for circumcising anyone under the age of 15 years, including EIMC, or a client of any age with immature genitaliaFootnote 2 [23, 24]. Prior to this decision, 10–14-year-old clients made up over 40% of PEPFAR’s annual VMMC volume.

While reporting into the NAERS is required, completeness or representativeness of the reporting is not known, and underreporting is suspected. Any estimates of program-wide NAE rates are limited by the unknown degree of underreporting and should be considered minimums. Thus, the NAERS is an important signal-generating process but is not exhaustive surveillance of all significant NAEs that occur in the context of PEPFAR VMMC programs. Differences in numbers between countries are likely due to differences in the number of VMMCs performed, and completeness of reporting, rather than overall differences in safety. Attempts to calculate NAE rates for comparison among programs based on these data could give misleading information and must acknowledge this limitation.

The increase in number of NAEs reported over time is likely due to increases in partner capacity to recognize and record NAEs, reporting compliance, and number of PEPFAR-supported VMMCs done each year. Efforts to increase reporting include reminders at regular meetings of PEPFAR country coordinators and at technical meetings, quarterly adverse event summaries shared with all PEPFAR teams, and assessment of reporting protocols during quality assurance assessments.