Validation and comparison of clinical prediction rules for invasive candidiasis in intensive care unit patients: a matched case-control study

Abstract

IntroductionDue to the increasing prevalence and severity of invasive candidiasis, investigators have developed clinical prediction rules to identify patients who may benefit from antifungal prophylaxis or early empiric therapy. The aims of this study were to validate and compare the Paphitou and Ostrosky-Zeichner clinical prediction rules in ICU patients in a 689-bed academic medical center.MethodsWe conducted a retrospective matched case-control study from May 2003 to June 2008 to evaluate the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of each rule. Cases included adults with ICU stays of at least four days and invasive candidiasis matched to three controls by age, gender and ICU admission date. The clinical prediction rules were applied to cases and controls via retrospective chart review to evaluate the success of the rules in predicting invasive candidiasis. Paphitou's rule included diabetes, total parenteral nutrition (TPN) and dialysis with or without antibiotics. Ostrosky-Zeichner's rule included antibiotics or central venous catheter plus at least two of the following: surgery, immunosuppression, TPN, dialysis, corticosteroids and pancreatitis. Conditional logistic regression was performed to evaluate the rules. Discriminative power was evaluated by area under the receiver operating characteristic curve (AUC ROC).ResultsA total of 352 patients were included (88 cases and 264 controls). The incidence of invasive candidiasis among adults with an ICU stay of at least four days was 2.3%. The prediction rules performed similarly, exhibiting low PPVs (0.041 to 0.054), high NPVs (0.983 to 0.990) and AUC ROCs (0.649 to 0.705). A new prediction rule (Nebraska Medical Center rule) was developed with PPVs, NPVs and AUC ROCs of 0.047, 0.994 and 0.770, respectively.ConclusionsBased on low PPVs and high NPVs, the rules are most useful for identifying patients who are not likely to develop invasive candidiasis, potentially preventing unnecessary antifungal use, optimizing patient ICU care and facilitating the design of forthcoming antifungal clinical trials.

Development and validation of a risk model for identification of non-neutropenic, critically ill adult patients at high risk of invasive Candida infection: the Fungal Infection Risk Evaluation (FIRE) Study

“…Playford et al 46 validated four clinical decision rules -the best rule from Ostrosky-Zeichner et al 27 and a subsequent revision to this, published in abstract form, and the two best rules from Pittet et al 43 -in a prospective cohort of 615 patients admitted for at least 72 hours to four multidisciplinary ICUs in Australia. Performance of the clinical prediction rules was worse than in the development data sets and the authors recommended that to identify a sufficiently high-risk population to consider for antifungal therapy would require a combination of the clinical risk factors from Ostrosky-Zeichner et al 27 with measures of colonisation from Pittet et al 43 Most recently, Hermsen et al 47 set out to validate the clinical decision rules of Paphitou et al 28 and Ostrosky-Zeichner et al 27 in a case-control study of 88 cases and 264 matched control subjects staying at least 4 days in a single multidisciplinary ICU in the USA. Rather than validate the rules as published, Hermsen et al 47 fitted new conditional logistic regression models using the risk factors from these rules, rendering their results incomparable with the original publications.…”

“…Performance of the clinical prediction rules was worse than in the development data sets and the authors recommended that to identify a sufficiently high-risk population to consider for antifungal therapy would require a combination of the clinical risk factors from Ostrosky-Zeichner et al 27 with measures of colonisation from Pittet et al 43 Most recently, Hermsen et al 47 set out to validate the clinical decision rules of Paphitou et al 28 and Ostrosky-Zeichner et al 27 in a case-control study of 88 cases and 264 matched control subjects staying at least 4 days in a single multidisciplinary ICU in the USA. Rather than validate the rules as published, Hermsen et al 47 fitted new conditional logistic regression models using the risk factors from these rules, rendering their results incomparable with the original publications. It is, however, worth noting that a number of the risk factors included in the rules (surgery, pancreatitis, haemodialysis and diabetes) were not subsequently found to be significantly associated with risk of IFD.…”

Karakteristik Klinis dan Prevalensi Pasien Kandidiasis Invasif di Rumah Sakit Cipto Mangunkusumo

“…[4][5][6][7] Selain insiden yang makin meningkat setiap tahunnya, beberapa studi di luar negeri juga menunjukkan bahwa KI juga menyebabkan tingginya angka mortalitas dan meningkatnya lama masa rawat, serta biaya perawatan. 6,8,9 Saat ini, belum ada studi di Indonesia yang meneliti tentang KI pada pasien dewasa. Studi yang pernah dilakukan di Indonesia yaitu terkait infeksi jamur invasif pada pasien neonatus oleh Wijayanto, dkk.…”

Invasive candidiasis in critical care setting, updated recommendations from “Invasive Fungal Infections-Clinical Forum”, Iran

“…The candida score of ≥ 3 is shown to predict IC with a sensitivity and a specificity of 81% and 74%, respectively [9] . These "risk factors" which are proposed as Candida Prediction Rules have been reported in many other studies [47][48][49][50][51][52][53] . In addition to Leon's Candida Score [47] and the Ostrosky-Zeichner [23] model, other models such as Agvald-Ohman et al [48] , Pittet et al [49] , Hermsen et al [51] , Paphitou et al [52] , and Dupont et al [53] tried to establish similar frameworks putting together risk factors which contribute to IC while assigning separate relative risk scores for each variable.…”

“…Some of the important differences in IC risk prediction models which are outlined in Figure 1 include the heterogeneity in the examined populations, non-similarity in the underlying disease severity, incidence of IC in centers where the investigations were carried out and the study end-points. It should be noted that most models were defined in surgical ICU populations [23,48,49,[51][52][53] . Intra-abdominal infections secondary to intestinal perforations and anastomotic leakage are also among the risk factors in patients who tend to mostly benefit from multi-center analysis on the prevalence of deep-seated mycosis in immunocompromised hosts in Tehran, Iran, Candida spp.…”