Clinical risk factors for postoperative nausea and vomiting (PONV) are well described, whereas genetic findings are conflicting.
OBJECTIVE
The aim of this study was to investigate a possible association of genetic variants and nongenetic variables with the incidence and severity of PONV.
DESIGN
A prospective observational study in two independent and different patient cohorts.
SETTING
Two independent patient cohorts differing in surgical procedures were enrolled in two tertiary care hospitals between 2008 and 2016.
PATIENTS
Consecutive patients of European origin undergoing elective surgery in two university hospitals. Clinical data were collected up to 24 h after surgery, and blood was drawn for genotyping. Of 2773 patients enrolled, 918 (Cohort A) and 1663 (Cohort B) with complete data sets were analysed.
MAIN OUTCOME MEASURE
Patients were allocated to one of three groups (No PONV, Intermediate PONV or Severe PONV) depending on the frequency of vomiting, the severity of nausea and the need for antiemetics. Clinical variables and 13 genetic variants of seven candidate genes were evaluated for association with these three phenotypes. The cohorts were analysed separately by ordinal logistic regression analysis, treating PONV as a dependent ordinal three-stage variable. Odds ratios (ORs) with 95% confidence intervals were calculated.
RESULTS
In Cohort A, the main predictors for PONV were female sex [OR (95% CI): 3.6 (2.7 to 4.8), P < 0.0001], nonsmoking status 1.8 (1.3 to 2.5), P < 0.001, the SS genotype (5-HTTLPR, rs4795541) of the promoter polymorphism in the serotonin transporter 1.5 (1.1 to 2.1), P = 0.019, and patient age 0.99 (0.98 to 0.99), P = 0.013. Analysis of Cohort B was consistent with these findings [5-HTTLPR: 1.8 (1.4 to 2.3), P < 0.00001]. Sex-specific regression models confirmed this 5-HTTLPR association in women and men.
CONCLUSION
In two independent cohorts, in addition to the well known clinical factors, a polymorphism of 5-HTTLPR in the serotonin transporter was independently associated with PONV. A possible evaluation of this biomarker to improve risk prediction within the scope of precision medicine should be considered.
