Hemoglobin Levels and Postoperative Venous Thromboembolism Following Metabolic and Bariatric Surgery: A Single-Center Retrospective Cohort Study

Background: Venous thromboembolism (VTE) is an uncommon but clinically important complication after metabolic and bariatric surgery (MBS). Although established clinical risk factors for postoperative VTE have been described, the relationship between preoperative hemoglobin (Hb) levels and postoperative VTE in MBS remains insufficiently characterized. This study aimed to evaluate the association between preoperative Hb levels and postoperative VTE and to describe related hematologic and clinical characteristics in a single-center MBS cohort. Methods: We conducted a single-center retrospective cohort study of 1414 eligible patients who underwent primary MBS at King Khalid University Hospital, Riyadh, Saudi Arabia, from January 2016 to December 2020. For this study, postoperative VTE denotes a broader composite endpoint comprising imaging-confirmed pulmonary embolism or portal-vein thrombosis within one year, rather than the conventional deep-vein thrombosis/pulmonary-embolism construct alone. Events were identified when clinically suspected; routine screening imaging was not performed. Descriptive statistics, Mann–Whitney U tests, and Fisher’s exact tests were used. Given the occurrence of only six postoperative VTE events, multivariable modeling was not conducted. Results: Among 1414 patients (mean age 34.5 years; 54.1% female; mean BMI 45.1 kg/m2), laparoscopic sleeve gastrectomy accounted for 93.6% of procedures. Six postoperative VTE events, as defined by the study’s broader composite endpoint, were documented (0.4%): four pulmonary emboli (0.28%) and two portal-vein thromboses (0.14%); no postoperative deep-vein thrombosis was documented. Three events occurred during the first postoperative week, two during weeks 2–3, and one at 2–3 months. Median preoperative Hb was similar in patients with and without an event (134.5 vs. 134.0 g/L; p = 0.500). On exploratory univariable analyses, previous VTE and higher MCV were associated with postoperative VTE events (Fisher’s exact p < 0.001 and Mann–Whitney U p = 0.032, respectively). Conclusions: Previous VTE was associated with postoperative VTE events on exploratory univariable analysis (Fisher’s exact p < 0.001). This finding is clinically coherent with previous VTE being an established postoperative thrombotic risk factor and supports its continued importance in perioperative risk assessment after MBS. Higher MCV was also associated with events on exploratory univariable analysis (Mann–Whitney U p = 0.032). Although MCV is not an established MBS-specific VTE risk marker, its association in this cohort provides a hypothesis-generating hematologic signal that merits evaluation in larger datasets.