Low sensitivity of the exit site scoring system in detecting exit site infections in peritoneal dialysis patients

Introduction: To evaluate the utility of the exit site scoring system in diagnosing exit site infections, we compared its scores to the well established exit site categories outlined by Twardowski and Prowant. Methods: We examined the exit sites of 50 patients on peritoneal dialysis. Each exit site was given a score ranging from zero to >= 4 according to the exit site scoring system. Exit sites were simultaneously classified into one of the exit site diagnostic categories outlined by Twardowski and Prowant. The sensitivity and specifity of the exit site scoring system in diagnosing exit site infection was then calculated in reference to the exit sites classified as infected according to Twardowski criteria. Results: The proportions of perfect, good, equivocal, and infected exit sites were 24%, 22%, 32%, and 22%, respectively. 21 exit sites had an exit site score of zero. Of those, 12 (57.1%) were perfect, 7 (33.3%) were good, 1 (4.8%) was equivocal, and 1 (4.8%) was actually infected. 21 exit sites had an exit site score ranging from 1 to 3 without purulent discharge. Of those, 4 (18.2%) were good, 15 (68.2%) were equivocal, and 3 (13.6%) were infected. The specifity of the exit site scoring system in diagnosing exit site infection was 100%, but its sensitivity was only 63.6%. Conclusion: The exit site scoring system is not a sensitive tool for diagnosing exit site infection mainly because it ignores evaluation of the sinus for granulation tissue. Exuberant granulation tissue in the sinus tract is not always accompanied by purulent drainage.