Risk Factors for and Outcomes of Catheter-Associated Peritonitis in Children: The SCOPE Collaborative

Discussion

In this large cohort of pediatric patients receiving chronic PD, one third ultimately developed one or more episodes of peritonitis. Rates of peritonitis were highest in children <2 years of age and lowest in those 13–17 years of age. As reported in prior studies, touch contamination and upward orientation of the catheter were identified as risk factors for infection (3). At the patient level, lower provider compliance with the follow-up care bundle was associated with a higher risk of peritonitis. This novel finding suggests that improved compliance with standardized care practices during a clinic encounter could reduce the rate of catheter-associated infections.

Quality improvement practices implemented by the SCOPE Collaborative have been shown to influence the rate of peritonitis. Neu et al. (7) previously reported that there was a reduction in the monthly annualized peritonitis rate from 0.63 pre-SCOPE Collaborative to 0.42 at 36 months post-SCOPE Collaborative as center-specific compliance with the follow-up bundle increased. That analysis focused on compliance and infection over time at the collaborative and center levels. This analysis sought to build on those findings by identifying risk factors for infection, including patient level compliance with the care bundles. Our findings reveal that provider compliance with the follow-up bundle at the patient level was associated with a significantly lower risk of peritonitis. When evaluating the individual components of the bundle, compliance with the review of hand washing, exit site care, and aseptic technique was significantly associated with a lower risk for peritonitis. Exit site scoring was not associated with a lower risk of peritonitis, but compliance with this bundle element was >96% in patients with and without a history of peritonitis. Interestingly, compliance with the demonstration and concept test every 6 months was not associated with the risk for peritonitis. Compliance with this bundle element was relatively low in both groups, and it must be emphasized that, for this analysis, follow-up compliance was measured at one visit only: the visit immediately before infection in the peritonitis group and at last follow-up in the group without peritonitis. In fact, analyses at the collaborative level suggest that the significant reduction in peritonitis rates over time was associated with increased compliance with the entire follow-up bundle, including the requirement for a repeat of the demonstration and concept test every 6 months (7). It will be important to repeat these analyses with additional follow-up time to more clearly evaluate associations between compliance with this specific care bundle and peritonitis risk.

It is also interesting to note that we were not able to show an association between compliance with the training bundle and the risk for peritonitis. The relatively low numbers of training events may explain the lack of a relationship, and a stronger effect may emerge as the numbers increase. It should be noted that the overall low compliance with the training bundle was because of low compliance with the home visit, and compliance with the other components of the bundle was high in both groups. Nevertheless, it is possible that the review of key aspects of catheter care and the dialysis procedure at each follow-up visit may minimize the effect of the initial training compliance beyond the first few weeks after that training. This is supported by the association between low compliance with review of these protocols and risk for peritonitis among all patients, including prevalent patients who would have received training on these bundles during a follow-up session rather than during their initial training, which occurred before the start of the SCOPE Collaborative.

Multiple reports have established an inverse correlation between rates of peritonitis and age (3). In this cohort, the highest rates of infection were seen in the youngest patients, although age was not a risk factor for peritonitis in the multivariable model. However, we did confirm several modifiable risk factors for peritonitis, including upward orientation of the catheter and a history of touch contamination. Consensus guidelines published by the ISPD in 2012 recommended placement of catheters with downward or lateral tunnel configuration to decrease the risk of infection (8). Despite these recommendations, 30 children in this study cohort had catheters placed with an upward–directed exit site, and more than one half of these children developed an infection. These data suggest that better adherence to the existing guideline on exit site orientation has the potential to reduce the frequency of peritonitis episodes in children. Interestingly, the presence of a gastrostomy tube at catheter insertion did not increase the risk of infection (9). Additional analyses to evaluate if risk for infection is associated with the timing of placement of the gastrostomy tube relative to placement of the PD catheter are planned.

The relative distribution of gram-positive, gram-negative, and culture-negative peritonitis in this cohort was similar to that of other recent reports (10–12). Common skin flora, including S. epidermidis, other coagulase–negative staphylococci, and Staphylococcus aureus, were the most common gram–positive pathogens isolated, consistent with the importance of touch contamination as a risk factor for infection. No pathogen was able to be identified in 25% of patients with peritonitis. The ISPD guidelines recommend that the rate of culture-negative peritonitis should not exceed 20% of episodes in any single center, with a goal of <10% (8). According to data from the International Pediatric Peritonitis Registry, the culture–negative peritonitis rate reported from 13 North American centers was only 11%, although there was substantial variability worldwide (12). Additional evaluation of the relatively high rate of culture-negative peritonitis is planned, including collecting center–specific culture techniques.

Fungal peritonitis has historically accounted for <5% of all episodes in children, but it is associated with significant morbidity, including technique failure (13). The rate of fungal peritonitis in the SCOPE Collaborative was higher than rates reported in the literature (3,12,14). Potential risk factors for fungal peritonitis, including prior antibiotic use and prior bacterial peritonitis, in addition to treatment and outcome data are currently being analyzed and will be published separately.

There are several limitations of this study that should be considered. Selection bias is a possible limitation, because centers voluntarily enroll patients and enter data in the SCOPE Collaborative database. However, the substantial number of centers and patients in the collaborative makes it likely that the data obtained are reflective of the pediatric PD population in general. Additionally, although the collaborative recommends that centers follow international guidelines for diagnosis and treatment of peritonitis (8), there may be variation in treatment of peritonitis by center, which may affect outcomes and hospitalization rates. Another limitation is that there was no central adjudication of infection events; peritonitis was reported by center but in accordance with the ISPD recommendations. Lastly, because the mission of the SCOPE Collaborative is to reduce infection rates by implementing quality initiatives, the rates of peritonitis and outcomes reported here may differ from centers that are not involved in the collaborative.

In summary, this study reveals that lower provider compliance with standardized practices for follow-up PD catheter care in the SCOPE Collaborative was associated with higher risk of peritonitis. In addition, orientation of the catheter exit site and touch contamination were identified as additional risk factors for peritonitis. These data support the hypotheses that repeated review of key aspects of PD catheter care and the aseptic technique required to perform the dialysis procedure and prevention strategies, such as avoidance of an upward orientation of the PD catheter exit site and prophylactic treatment of touch contamination episodes, may have the potential to reduce PD-associated peritonitis. The results from the SCOPE Collaborative may provide direction to guide best practice development for the care of children on chronic PD.