We would like to report on a randomised, double-blind controlled trial we undertook in our University Hospital, Department of Anaesthesiology and Intensive Care Medicine in Berlin, Germany.
The primary aim of this trial was to investigate the effects on mHLA-DR (monocytic human leucocyte antigen) expression and postoperative infection rates of preoperative vaccination in cancer patients undergoing surgery of the upper aerodigestive tract.
In a previous pilot study, our research group found an increase of mHLA-DR expression after influenza vaccination compared with placebo in patients with cancer of the upper aero-digestive tract without subsequent surgery.1
Ethics approval for this study was provided by the local ethics committee (Ethics board Charité–University Medicine Berlin, Campus Mitte, chairperson: Prof. Uebelhack), on 9 March 2005 (ref: EA 1/048/04). The study evaluated consecutive patients scheduled for surgery on upper aero digestive tract cancer (neck dissection or upper oesophageal resection) between November 2005 and July 2008. Patients were excluded if less than 18 years old, pregnant or lactating, were of American Society of Anesthesiologists (ASA) classification IV or V, had any diagnosed infection in the previous 14 days, were HIV positive, had liver cirrhosis (Child B or C), had undergone organ transplantation, were immune suppressed, had been treated with corticosteroids in the previous 4 weeks, had a recent myocardial infarction (less than 8 weeks previously), had diabetes mellitus with sequelae, had undergone a splenectomy, had a hypoproteinaemia, had a known incompatibility or allergy against any component of the vaccine or had a significant psychiatric illness.
After obtaining written informed consent, patients were randomised to one of the two treatment groups: vaccination group or the placebo group stratified to the type of surgery.
The first vaccination was performed on the day before surgery at 07:00 p.m.; the second vaccination was performed on the day of surgery at 7 : 00 a.m.
Four blood samples for analysis of mHLA-DR expression were taken from each patient: immediately before the first vaccination, immediately before the second vaccination, and on the first and third postoperative days (POD). Vaccination was performed with the haemagglutinin antigens of the influenza virus (Mutagrip; Sanofi Pasteur MSD, Lyon, France) for the northern hemisphere. The nonvaccinated group of patients received 0.5 ml 0.9% normal saline subcutaneously.
Statistical analyses of metric parameters were performed by nonparametric tests such as (exact) Mann–Whitney U test (for independent samples). Nominal-scaled clinical data were analysed using (exact) Chi-square tests. Differences between treatments with reference to baseline were tested with both univariate and multivariate nonparametric analysis of covariance for repeated measures. All results for metrically scaled data are given as median and interquartile range [25 to 75%].
The intention-to-treat population consisted of 66 patients (33 per group). Baseline characteristics did not differ between the two groups except for the frequency of alcohol use disorders and known allergies: these were more frequent in the vaccination group.
Both groups showed postoperative immunodepression, documented by decreased mHLA-DR expression on day 1 and 3 after surgery, but there was no significant difference between vaccinated and nonvaccinated patients. The proportion of patients with mHLA-DR expression of less than 10 000 antibodies per cell was 27% in the placebo group vs. 19% in the vaccination group (P = 0.218).
Cumulative incidence of postoperative infections is shown in Fig. 1. Overall, 10 days after surgery, infection rate did not differ between the groups. However, until POD 5, there was a slight tendency for decreased infection rates in the vaccination group (P = 0.190), whereas between POD 6 and POD 10, the incidence of infections increased in vaccinated patients compared with the nonvaccinated patients (P = 0.016). The following infections occurred (according to the centers for disease control and prevention definitions for nosocomial infections): surgical wound infection, pneumonia, bloodstream infection (with sepsis) and urinary tract infection.
One patient in each group died. Median duration of ICU stay in the vaccination group was 2.2 [0.8 to 6.7] days vs. 3.3 [0.7 to 4.9] days in the nonvaccination group (P = 0.807). Mean hospital stay in the vaccination group was 12.5 [9.3 to 24.0] days vs. 14.0 [11.8 to 22.3] days in the nonvaccination group (P = 0.305).
Our results suggest that postoperative immune suppression cannot be avoided by preoperative vaccination measured by mHLA-DR expression on monocytes.
Other studies that addressed restoring monocyte function and mHLA-DR expression as well as cell-mediated immune response by influenza vaccination were performed only in nonsurgical patients.2,3 Haining et al.2 demonstrated an increased mHLA-DR expression after influenza vaccination in children after haemopoietic stem cell transplantations. The study of Holvast et al.3 revealed an increase of interferon (IFN)-gamma after influenza vaccination in patients with systemic lupus erythematosus (SLE) and healthy control individuals. These effects were not seen after surgery in our study. In addition, recent evidence looking at leucocyte gene expression in patients with trauma and burns suggest that the pro and anti-inflammatory responses occur simultaneously.4 This is in contrast to the paradigm that trauma/surgery leads to a pro-inflammatory response phase followed by an anti-inflammatory response phase on which the rationale of this trial was based.
The postoperative incidence of infections did not differ between vaccinated and nonvaccinated patients from POD 0 to POD 5, which is consistent with the lack of vaccination effect on cell-mediated immune response. From POD 6 to POD 10, the cumulative incidence of infections was higher in vaccinated patients than in nonvaccinated patients. This result is in contrast to our hypothesis that preoperative vaccination can reduce the postoperative incidence of infections. It was not an issue of the safety analysis, as this result became apparent after un-blinding all patients during final data analysis.
As the trial is underpowered, the findings have to be interpreted with caution. Due to the small sample size, it was not possible to consider possible different effects between the surgical procedures (i.e. neck dissection and oesophageal resection) nor was it possible to consider the effects of additional confounders such as the type of anaesthesia, duration of surgery or blood loss (although duration of surgery or blood loss did not differ between the groups).
Our results suggest that the vaccination of patients before surgery does not enhance HLA-DR expression or reduce postoperative infections. Whether earlier vaccination would be effective is unknown. No serious adverse reactions were observed.
Acknowledgements relating to this article
Assistance with the study: we are grateful to Claudia Nitschke, Anja Baethge, Jana Kindermann, Annabell Hijazi, Pascal Kucik, Birgit Rückershausen and Sarah Keller for data and blood sample collection.
Financial support and sponsorship: the trial was financially supported by inner university grants and DFG grant 432 (German Research Foundation).
Conflicts of interest: none.
ISRCTN Registration Number: ISRCTN16108096.