Nail Clubbing: A Dermatologic Window into Underlying Systemic Disease – an All of Us Study

Nail clubbing is defined as increased nail bed soft tissue volume resulting in obliteration of the angle between the proximal nail fold and the nail bed [1]. It is theorized that chronic cyanosis causes a neurocirculatory reflex leading to increased blood flow through digital arteriovenous shunts resulting in tissue hypertrophy and hyperplasia. Prostaglandins, ferritin, and vascular endothelial growth factors have been implicated in its pathogenesis [2]. The presentation of nail clubbing varies with severity. Initially, it is characterized by periungual erythema and nail bed softening with a spongy feel. In advanced stages, there is distal phalanx thickening resulting in a bulbous appearance of the digit, a shiny appearance of the nail and periungual skin, nail fold erythema, longitudinal nail plate ridging, and increased vascularity leading to a lilac hue of the nail bed [3‒5].

Nail clubbing can manifest independently or more commonly as part of a secondary disease process. While respiratory disease is frequently implicated, with 30% of patients having pulmonary disease, cardiovascular diseases, including congenital cyanotic heart disease; gastrointestinal diseases, including inflammatory bowel disease; endocrine disorders, including Graves’ disease; and rarely hereditary clubbing have also been associated with digital clubbing [1, 6‒11]. Studies describing digital clubbing in patients without pulmonary pathology are limited [12, 13]. Furthermore, only one study analyzed the sociodemographic characteristics of patients with a broad range of non-pulmonary conditions associated with nail clubbing [14]. We aimed to describe non-pulmonary medical conditions associated with nail clubbing using a cross-sectional approach to enable physicians to improve diagnosis of nail clubbing and make appropriate screening referrals.

We conducted a cross-sectional analysis of the All of Us (AoU) Research Program, a National Institute of Health initiative dedicated to enrolling diverse and underrepresented populations in biomedical research. Nail clubbing cases were identified by ICD-10-CM code R68.3 and/or SNOMED code 4373005. Patients were stratified by pulmonary pathology reported in the AoU electronic health record (EHR) including “chronic obstructive lung disease,” “disorder of lung,” “malignant tumor of lung,” “mesothelioma (malignant, clinical disorder),” and “pulmonary hypertension.” Patient characteristics and demographic information of nail clubbing patients with one or more pulmonary disease (cases) and nail clubbing patients without pulmonary disease (controls) were extracted from EHR and survey data. Comparisons of demographics and clinical characteristics between cases and controls were analyzed with Fisher’s exact test or Pearson’s χ² test for categorical variables. Continuous variables were analyzed with independent two-sample t tests. Odds ratios of nail clubbing diagnosis were analyzed with multivariate logistic regression and adjusted for sociodemographic characteristics.

Of the 407,333 adults in the AoU Registered Tier Dataset v7 (enrolled between May 2017 and June 2022), 266,612 (65.5%) had analyzable EHR data. In total, 85 participants had a diagnosis of nail clubbing (overall prevalence 0.03%), of which 63.53% had a pulmonary disease versus 36.47% of controls without documented pulmonary pathology. Mean age for cases and controls was similar (61.12 years and 65.27 years, respectively, p = 0.99) (Table 1). Overall, 57.41% of cases were female versus 35.48% of controls (p = 0.07) (Table 1). Patients with nail clubbing were more often white (42.35%) than black (37.65%) or another race and similar between cases and controls (p = 0.79). The majority of the cohort was of non-Hispanic ethnicity (82.35%), which was similar between cases and controls (p = 0.77) (Table 1). The highest level of education was comparable for both groups, with 35.29% of patients completing “some college” (p = 0.11) (Table 1). There was no difference in tobacco (p = 0.54) or alcohol consumption habits (p = 1.0) between cases and controls (Table 1). Overall, across both cases and controls, approximately 22% of patients had chronic liver disease, 17% had hypothyroidism, 8% had HIV infection, 5% had congenital heart disease, and 5% had Graves’ disease or hyperthyroidism (Table 2). Male versus female patients with nail clubbing had decreased odds of having concurrent respiratory disease diagnosis (odds ratio, 0.37; 95% confidence interval [CI], 0.14–0.92, p = 0.03) (Table 3).

We found that more than a third of cases in our cohort had nail clubbing associated with non-pulmonary pathologies. Similarly, a 2022 systematic review and meta-analysis of 142 studies, including 3,790 patients, reported pooled prevalence of digital clubbing of 33.4% (95% CI, 16.6–52.8), 31.3% (95% CI, 22.4–41.1), 27% (95% CI, 9.4–49.5), and 22.8% (95% CI, 10.8–37.6) in subjects with intestinal diseases, interstitial lung diseases, infective endocarditis, and hepatic diseases, respectively [14]. Notably, the sex distribution of patients with nail clubbing categorized by underlying systemic disease has not previously been described. We found that male patients with nail clubbing were less likely to have pulmonary pathology than female patients. Taken together, our findings and others emphasize the importance of a workup of non-pulmonary pathologies as a possible cause of digital clubbing.

Our findings suggest that multiple systemic diseases are associated with nail clubbing. For this reason, it is imperative to approach the workup of patients presenting with nail clubbing with a broad lens to assure no underlying pathology is overlooked. An algorithmic approach to caring for nail clubbing patients includes a comprehensive history including review of systems. Particular attention is paid to questions that would reveal symptoms related to the medical comorbidities associated with nail clubbing including pulmonary disease (shortness of breath, cough, wheezing), heart disease (chest pain, palpitations), chronic liver disease (yellowing of the skin or eyes, right upper quadrant pain, steatorrhea), inflammatory bowel disease (abdominal pain, hematochezia), hyperthyroidism (goiter or history of goiter, heat intolerance, tremor, weight loss), hypothyroidism (constipation, cold intolerance, weight gain), or hematologic pathology (easy bruising, lymphadenopathy, clotting issues). A full dermatologic examination is necessary paying close attention to findings that may reveal underlying systemic illness, for example, jaundice, spider nevi, and leukonychia associated with chronic liver disease, or alopecia, brittle nails, and acquired ichthyosis associated with hypothyroidism [15, 16]. Workup includes referral to a primary care physician or specialist(s) guided by the history and physical examination.

Limitations of our study include reliance on EHR data and AoU data dissemination guidelines which stipulate that values less than or equal to 20 are reported as “≤20.” In this study, all statistical analyses were performed with raw data; however, Tables 1 and 2 report redacted “≤20” for corresponding values. Additionally, there was an overrepresentation of black participants and an underrepresentation of Asian participants in the AoU database compared to the general US population. Despite these limitations, this large database of a diverse patient population is a useful tool to better understand dermatologic pathologies. The small sample size of nail clubbing patients in AoU suggests a need for increased patient recruitment so that the prevalence of dermatologic diseases in the database better reflects the US population.

In conclusion, greater than one-third of patients in our cohort had nail clubbing associated with a non-pulmonary systemic disease and a significant proportion were male. Consideration of a broad differential of both pulmonary and non-pulmonary pathologies associated with nail clubbing is needed to initiate workup and make appropriate screening referrals. Our study highlights the need for further research with larger cohorts to further characterize demographic differences in the systemic pathologies that underlie nail clubbing. We also elucidate the important role that dermatologists can play in the diagnostic workup of chronic systemic diseases with dermatologic manifestations.

We gratefully thank All of Us participants for their contributions, without whom this research would not have been possible. We also thank the National Institutes of Health’s All of Us Research Program for making available the participant data examined in this study.

This study protocol was reviewed and approved by Weill Cornell Medicine IRB, Approval No. 22-10025306 “Skin and Nail Demographics using All of Us Database.” This study did not include human participants.

Ms. Desir has no conflicts of interest. Dr. Lipner has served as a consultant for Ortho Dermatologics, BelleTorus Corporation, Moberg Pharmaceuticals, and Hoth Therapeutics.

This study was not funded.

All authors have contributed substantially to the production of this manuscript. Noelle Desir performed data collection, statistical analysis, and manuscript drafting. Dr. Lipner performed manuscript drafting and is the corresponding author.

Study data are available at the All of Us Researcher Workbench and can be accessed upon completion of the following: (1) create Researcher Workbench account at https://www.researchallofus.org/register/ (your institution must have a Data Use and Registration Agreement [DURA] in place with All of Us), (2) verify identity through Login.gov or ID.me, (3) complete mandatory training required to access the Controlled Tier dataset, and (4) sign the data user code of conduct (DUCC).