UNDERDIAGNOSIS OF CHRONIC KIDNEY DISEASE IN THE NURSING HOME POPULATION

Abstract

To the Editor: Chronic kidney disease (CKD) is present in more than 12% of Americans aged 65 and older.1 In the guidelines from the National Kidney Foundation (NKF) Kidney Disease Outcome Quality Initiative, age 60 and older is indeed considered to be a major risk factor for CKD.2 Among various complications, CKD appears to independently predict mortality and cardiovascular disease.3–5 Readily available formulas for calculating glomerular filtration rates (GFRs), along with a staging system and CKD stage-dependent therapeutic guidelines, have simplified the ability to identify CKD, categorize its severity, and implement appropriate treatment. Nevertheless, several reports suggest that CKD is underdiagnosed and undertreated.1,2,6 The purpose of this study was to explore how frequently physicians of elderly nursing home residents, who have CKD based on NKF criteria, address this diagnosis. After institutional review board approval was obtained, a retrospective chart review was conducted of all long-term residents in a 672-bed facility aged 60 and older who had resided there for at least 6 months and whose records included at least two serum creatinine levels drawn at least 90 days apart from each other. Each subject's monthly physician progress notes over the previous 6 months were reviewed to determine whether a diagnosis of CKD was recorded. The Modification of Diet in Renal Disease (MDRD) and Cockcroft-Gault (C-G) formulas were used to estimate subjects' GFRs,7,8 with values less than 60 mL/min per 1.73 m2 of body surface area regarded as positive screening tests for CKD. Serum creatinine values recommended for identification of CKD (≥1.5 mg/dL for men, ≥1.3 mg/dL for women) were also used as screening criteria for CKD.9 Two hundred eighty patients met the criteria. Of those with CKD according to MDRD and C-G criteria, a diagnosis of CKD was not noted in 62% and 82%, respectively, of the charts. Of women with CKD according to MDRD and C-G criteria, 70.4% and 87%, respectively, had no notation of CKD in their charts. A diagnosis of CKD was omitted less often in men with CKD (35.3% and 62.9% when identified according to MDRD and C-G, respectively). A diagnosis of CKD was also frequently omitted from charts of patients with CKD based on aforementioned sex-based serum creatinine values as well. Using logistic regression analysis (P=.02), when controlling for age, sex was found to significantly affect the likelihood of CKD being recognized. Men had only 0.25 odds of underdiagnosis of CKD when compared with women (P=.049). When sex was controlled for, there was no significant relationship between age groups and underdiagnosis using the MDRD equation (Table 1). Using patients with CKD according to C-G, a similar effect of sex was observed using logistic regression (P=.01), but when sex was controlled for, patients who were aged 71 to 80 had only a 0.23 odds of underdiagnosis when compared with those aged 90 and older (P=.02). Delayed recognition and therapy of CKD may predispose patients to adverse outcomes, and these data suggest that CKD may be substantially underdiagnosed in the elderly nursing home population. Although CKD was addressed in only a minority of patients in whom it was evident using GFR estimations, the presence of CKD was documented more frequently when using the creatinine-based parameters described previously.9 Although there are a number of potential explanations why such a difference was observed, it may simply be that an overtly high serum creatinine level will be more likely to draw the physician's attention than a relatively “normal” appearing serum creatinine level that nevertheless corresponds with a diminished GFR that has not been calculated. The NKF guidelines not only recommend use of GFR estimation equations, but also expressly declare use of serum creatinine alone not to be optimal in assessment of kidney function.2,9,10 Although GFR calculations and serum creatinine have limitations, these data nevertheless suggest substantial underdiagnosis of CKD, even with serum creatinine levels above 1.4 mg/dL. In summary, despite well-established criteria for the diagnosis of CKD, including simple methods to estimate GFR, CKD appears to be underdetected within the nursing home setting, potentially placing this community at risk for costly, avoidable outcomes. This study underlines the effect of age and sex on misdiagnosis of CKD. Further studies will be required to identify variables accounting for this low rate of identification and to devise strategies to facilitate recognition and treatment of CKD in elderly patients. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this letter. Author Contributions: All authors were involved in the concept, design, acquisition of data, analysis and interpretation of data, and preparation of manuscript. Sponsor's Role: None.