E. Mimouni1, J.E. Passman1, M. Alexis2, S.P. Ginzberg1, R. Kelz1, H. Wachtel1 1Hospital of the University of Pennsylvania, Department Of Surgery, Philadelphia, PENNSYLVANIA, USA 2Perelman School of Medicine, Philadelphia, PENNSYLVANIA, USA
Introduction:
The need for thyroid hormone therapy (LT4) after partial thyroidectomy (PT) varies with clinical factors, including preoperative thyroid stimulating hormone (TSH) level and diagnosis. The goal of this study was to develop a practical, easy-to-use clinical risk score to predict need for LT4 after PT.
Methods:
This is a retrospective cohort study of patients who underwent PT (defined as CPT codes 60220 or 60225) between 2013-2020 at a single institution. Patients on preoperative LT4 were excluded. Patient clinical data were extracted from the electronic health record. Elixhauser comorbidity scores were calculated using the ICD-10 codes. The primary outcome was post-operative LT4 therapy. Univariable and multivariable logistic regression were used to evaluate the relationship between clinical covariates and LT4 therapy. Covariates significantly associated with LT4 therapy were incorporated into a weighted risk score, with points assigned by β-coefficients as follows: β<0.5 = 1 point; 0.5≤β<1 = 2 points; 1≤β<1.5 = 3 points; β>1.5 = 4 points as previously published. Performance characteristics for the risk score were assessed.
Results:
Of 1165 patients, 906 (78%) were female, and the median age was 51 years (IQR:39-62). The median preoperative TSH level was 1.4 uIU/mL (IQR:0.9-2.0) and 88 patients (7.6%) had a pre-existing diagnosis of hypothyroidism but were not on LT4. Surgical pathology was cancer in 49% (n=572) with papillary thyroid cancer being the most common histopathology (n=512). In total, 38% (n=441) of patients required LT4 supplement after PT. On multivariable logistic regression, higher TSH (OR:2.03, 95% CI:1.59 – 2.62, p<0.001), higher Elixhauser comorbidity index (OR:1.13, 95% CI:1.03 – 1.23, <0.001), non-Black race (OR:3.69; 95% CI:1.99-7.15; p<0.001), and cancer diagnosis (OR: 2.14, 95% CI:1.40 – 3.29, p<0.001) were significantly associated with higher likelihood of LT4 after PT (as shown in Table) and were incorporated into a clinical risk score. Rates of LT4 use increased with risk score (0-1 points: 15%; 2-4 points: 35%; ≥5 points: 51%). The risk score had a negative predictive value (NPV) of 71% for LT4 therapy, with a Harrell’s C=0.75.
Conclusion:
Higher TSH, cancer diagnosis, race and Elixhauser comorbidity index were significantly associated with a higher likelihood of requiring LT4 after PT. A weighted risk score integrating these four clinical features had a good predictive value for LT4 therapy. This simple, easy-to-derive risk score can be used to counsel patients about likelihood of needing LT4 after PT, and may help guide clinical discussions about the extent of surgery in patients with thyroid nodules.