The Role of Inpatient Dermatology Consultations
Inpatient dermatology has been shown to have profound effects on the care of hospitalized patients. However, dermatology consultations remain an underutilized resource. The purpose of this study was to demonstrate how dermatology affects the hospitalization of inpatients while highlighting the breadth of services provided. This cross-sectional retrospective study included all inpatient dermatology consultations completed at a large tertiary-care facility in an urban setting. It aimed to investigate the reasons for consultation, as well as the effects on diagnosis, management, disposition, and cutaneous condition by time of discharge. This study provides evidence supporting the integration of the dermatologist into the care of hospitalized patients by illuminating lesser-known areas of impact.
Practice Points
- Inpatient dermatologists fill knowledge gaps that often alter the diagnosis, management, and hospital course of hospitalized patients.
- Several medical specialties benefit from niche expertise of inpatient dermatologists specific to their patient population.
- Integration of inpatient dermatology consultations can prevent unnecessary hospital admissions and medication administration.
Impact on Diagnosis—Fifty-six percent of all consultations resulted in a change in diagnosis. When dermatology was consulted specifically to assist in the diagnosis of a patient (195 consultations), the working diagnosis of the primary team was changed 69% of the time. Thirty-five of these consultation requests had no preliminary diagnosis, and the primary team listed the working diagnosis as either rash or a morphologic description of the lesion(s). Sixty-three percent of suspected drug eruptions ended with a diagnosis of a form of drug eruption, while 20% of consultations for suspected cellulitis or bacterial infections were confirmed to be cellulitis or soft tissue infections.
Impact on Management—Regardless of the reason for the consultation, most consultations (86%) resulted in a change in management. The remaining 14% consisted of FBSEs with benign findings; cases of cutaneous metastases and leukemia cutis managed by oncology; as well as select cases of purpura fulminans, postfebrile desquamation, and postinflammatory hyperpigmentation.
Changes in management included alterations in medications, requests for additional laboratory work or imaging, additional consultation requests, biopsies, or specific wound care instructions. Seventy-five percent of all consultations were given specific medication recommendations by dermatology. Most (61%) were recommended to be given a topical steroid, antibiotic, or both. However, 45% of all consultations were recommended to initiate a systemic medication, most commonly antihistamines, antibiotics, steroids, antivirals, or immunomodulators. Dermatology recommended discontinuing specific medications in 16% of all consultations, with antibiotics being the most frequent culprit (17 antibiotics discontinued), owing to drug eruptions or misdiagnosed infections. Vancomycin, piperacillin-tazobactam, and trimethoprim-sulfamethoxazole were the most frequently discontinued antibiotics.
Dermatology was consulted for assistance in management of previously diagnosed cutaneous conditions 56 times (22% of all consultations), often regarding complicated cases of hidradenitis suppurativa (9 cases), pyoderma gangrenosum (5 cases), bullous pemphigoid (4 cases), or erythroderma (4 cases). Most of these cases required a single dermatology encounter to provide recommendations (71%), and 21% required 1 additional follow-up. Sixty-three percent of patients consulted for management assistance were noted to have improvement in their cutaneous condition by time of discharge, as documented by the primary provider in the medical record.
Twenty-eight percent of all consultations required at least 1 biopsy. Seventy-two percent of all biopsies were consistent with the dermatologist’s working diagnosis or highest-ranked differential diagnosis, and 16% of biopsy results were consistent with the second- or third-ranked diagnosis. The primary teams requested a biopsy 38 times to assist in diagnosis, as documented in the progress note or consultation request. Only 21 of these consultations (55% of requests) received at least 1 biopsy, as the remaining consultations did not require a biopsy to establish a diagnosis. The most common final diagnoses of consultations receiving biopsies included drug eruptions (5), leukemia cutis (4), vasculopathies (4), vasculitis (4), and calciphylaxis (3).
