Inadequate Hepatitis B Vaccination and Post-Exposure Evaluation Among Transplant Surgeons

Abstract

Objectives:

To identify the proportion of U.S. transplant surgeons who are adequately vaccinated against hepatitis B virus (HBV), identify characteristics associated with inadequate vaccination, and assess the proportion who had been evaluated for immunization following potential HBV exposures.

Summary Background Data:

It is unknown what proportion of transplant surgeons are appropriately vaccinated against HBV or evaluated for immunization following operative exposures.

Methods:

We mailed questionnaires and to all active U.S. transplant surgeons. We compared demographic characteristics of responders and nonresponders to evaluate the potential for nonresponse bias.

Results:

Of 619 eligible respondents, 347 (56.1%) returned completed questionnaires. Of the 311 surgeons for whom HBV vaccination was indicated (all surgeons with neither a prior history of HBV infection nor a prior adverse reaction to the vaccine itself), 70 (22.5%; 95% confidence interval [CI], 18.0–27.6%) received fewer than the recommended 3 injections. Surgeon characteristics associated with inadequate vaccination included length of clinical practice (odds ratio [OR], 1.5 per 10-year increment in duration of practice; 95% CI, 1.1–2.2), increased fear of infection (OR, 1.2 for each unit increase in fear out of 10; 95% CI, 1.1–1.4), and lack of recent testing for HBV infection (OR, 2.0; 95% CI, 1.1–3.8). Of the 94 surgeons (27.3%) reporting at least one needle-stick exposure while operating on an HBV-infected patient, 14 (14.9%) were inadequately vaccinated; of these 14, only 5 (35.7%) sought appropriate serologic testing and counseling for active immunization. Surgeons underestimated both the risks of percutaneous exposure while operating, and of becoming infected with HBV if exposed.

Conclusions:

Many transplant surgeons are inadequately vaccinated against HBV and fail to seek evaluation following possible exposures. Underestimation of the risks of HBV exposure and transmission may relate to these failures. Requiring documentation of HBV vaccination and immunity to maintain operating room privileges may protect surgeons, their patients, and operating room staff.

Transplant surgeons are at increased risk for hepatitis B virus infection. In a survey of all U.S. transplant surgeons, we found that substantial proportions both fail to be fully vaccinated against hepatitis B and fail to seek appropriate prophylaxis following potential intraoperative exposures. We recommend that, to protect patients, surgeons and other operating room staff be required to document adequate vaccination and immunity against hepatitis B virus.

Recent research and position papers have focused attention on surgeons’ intraoperative exposures to the hepatitis B, hepatitis C, and human immunodeficiency viruses.^1–10 However, surgeons’ perceptions of the probabilities of transmission of these viruses remain inaccurate,¹¹ and their utilization of several recommended prophylactic interventions remains incomplete.^11–16

Surveys conducted in the early 1990s showed that many surgeons are not fully vaccinated against the hepatitis B virus (HBV).^12–14 However, these studies were conducted before or shortly after implementation of the Occupational Safety and Health Administration's mandate that all United States healthcare workers be offered HBV vaccination.¹⁷ It is not known whether vaccination rates have improved subsequently.

There is also limited evidence regarding the characteristics of surgeons who fail to use recommended prophylactic interventions, and whether incomplete vaccination stems from erroneous perceptions of transmission risk, vaccination benefit, or from other factors. As a result, we neither know whom to target with interventions to promote vaccination nor what information to include in such interventions to promote behavior change.

In this study, we sought to quantify the proportion of United States transplant surgeons who were inadequately vaccinated against HBV in 2003. We also evaluated the proportion of surgeons who were evaluated for active HBV immunization following suspected intraoperative exposures. We sought to identify surgeon characteristics and risk estimates associated with inadequate utilization of these risk-reduction strategies. We chose to evaluate transplant surgeons because the characteristics of their patients and complexities oftheir procedures place them at highest risk for HBV exposure.¹²

METHODS

Our survey strategy has been described previously.¹⁸ Briefly, we surveyed all active U.S. transplant surgeons, excluding collaborators on this study (A.S. and P.S.). In January 2003, we mailed each of these 625 surgeons a 3-page questionnaire, a cover letter signed by the authors, a self-addressed stamped return envelope, and a $10 bill as an incentive for completion.¹⁹ After 3 weeks, we sent a second copy of the questionnaire to all surgeons who had not yet responded. We calculated the response rate as the number of surgeons returning questionnaires with more than 80% of items completed within 6 weeks of the second mailing, divided by the effective sample size.^19,20

Questionnaire

We asked surgeons to indicate the number of injections of HBV vaccine they had ever received. We considered responses of 3 or more injections to indicate adequate vaccination. Surgeons who had not been vaccinated were asked to indicate why they had not done so. We considered responses that the surgeon was immune based on prior infection, or allergic to the vaccine, to be legitimate indications for not being vaccinated. We also evaluated the number of times within the past 2 years that surgeons had operated on a patient known or suspected to be infected with HBV, the number of percutaneous exposures they had sustained while operating on such patients, and the number of times they sought postexposure evaluation following these exposures.

We asked surgeons to estimate the following: 1) the probability of having a percutaneous exposure to HBV while performing a transplant on an infected patient, 2) the probability of viral transmission if they were intraoperatively exposed to HBV-infected blood, and 3) the extent to which they would fear becoming infected with HBV while performing their most common transplant on an HBV-infected patient (on a scale from 1, representing “no fear at all,” to 10, representing “the most fear I've ever had”).

Analyses

Our primary outcome variable was adequacy of HBV vaccination, and we aimed to identify surgeon characteristics or beliefs associated with this outcome. A priori, we identified the variables listed in Table 1 as potential predictors of vaccination adequacy. We first evaluated bivariate associations between these variables and vaccination adequacy. We used χ² tests, 2-sample t tests, and Wilcoxon rank-sum tests for binomially, normally, and non-normally distributed data, respectively. We then forced into a multivariable logistic regression model all variables related to vaccination adequacy at P < 0.15 in these bivariate analyses. We used Stata 7.0 software (Stata Corp., College Station, TX) for all analyses.

TABLE 1. Characteristics of Transplant Surgeons by Hepatitis B Vaccination Status

Although this study was explicitly powered for an outcome reported separately,¹⁸ we estimate that we had 83% power to detect any binary characteristic of surgeons associated with an odds ratio of 2.0 for being inadequately vaccinated. This estimate is based on the anticipated receipt of 350 completed questionnaires, use of a two-sided significance level of 0.05, and allowance for up to a 2:1 ratio in the distribution of binary characteristics of surgeons.

We investigated a secondary outcome of whether surgeons had appropriately sought evaluation for prophylactic treatment with HBV vaccination, hepatitis B immune globulin, or both, after a suspected exposure to HBV-infected blood. We defined adequacy of postexposure evaluation conservatively by requiring only that surgeons sought appropriate evaluation following at least one of their suspected exposures. Because unique circumstances may have influenced each surgeon's experiences with subsequent needle-stick exposures, we did not require surgeons to have used postexposure care following each exposure. Surgeons who reported no exposures to HBV were excluded from this analysis.

To check the internal consistency of surgeons’ responses regarding postexposure evaluation and treatment, we used χ² tests to investigate associations between surgeons’ reports of seeking postexposure evaluation and their reports of having been tested for each virus within the past 3 years. If surgeons properly understood the question and responded accurately, those who indicated use of postexposure care within the last 2 years should also indicate having been tested for the virus because such care routinely involves testing for active infection.

The University of Pennsylvania Institutional Review Board approved this study.

RESULTS

Of the 619 eligible respondents, 347 (56.1%) returned completed questionnaires. To evaluate the potential for nonresponse bias, we compared the demographic and practice related characteristics between responders and nonresponders. As reported previously,¹⁸ we found no differences between responders and nonresponders in their sex, practice setting, or major professional activity (office-based practice vs. full-time hospital practice, teaching, research, or other). Responders were younger than nonresponders (mean age, 48 vs. 51 years, P < 0.0001). Transplant surgeons whose primary board certification was in urology were less likely to respond than those whose primary certification was in general or thoracic surgery (18% vs. 59%, P < 0.0001).

HBV Vaccination Adequacy

A total of 344 respondents (99.1%) indicated the number of vaccine injections they had received; 33 surgeons provided legitimate reasons for not having been vaccinated. Of the remaining 311 surgeons, 70 (22.5%, 95% confidence interval [CI], 18.0%–27.6%) were inadequately vaccinated. The characteristics of surgeons who were adequately and inadequately vaccinated against HBV are shown in Table 1.

The multivariable regression model revealed that years of practice (odds ratio [OR], 1.5 per 10-year increment in duration of practice; 95% CI, 1.1–2.2), and increased fear of becoming infected with HBV (OR, 1.2 for each unit increase in fear, out of 10; 95% CI, 1.1–1.4), were each associated with a greater odds of being inadequately vaccinated. In addition, surgeons who had not been tested for HBV infection within the past 3 years had greater odds of being inadequately vaccinated than surgeons who had been tested (OR, 2.0; 95% CI, 1.1–3.8).

Adequacy of HBV Postexposure Evaluation

Overall, 94 surgeons (27.3%) reported having sustained at least one needle-stick exposure while operating on an HBV-infected patient within the past 2 years. Of these, 41 (43.6%) reported having sought postexposure evaluation for possible receipt of HBV vaccination or immune globulin. Of the 53 surgeons who failed to seek postexposure evaluation, 9 were inadequately vaccinated and thus at maximal risk for infection.

No characteristics of surgeons were independently associated with their probability of seeking appropriate evaluation, but the small number of surgeons reporting exposures limits the power to detect such associations. All 41 surgeons (100%) who reported using postexposure care also reported being tested for HBV within the past 3 years, a significantly greater percentage than among surgeons who reported not using postexposure care following exposures (66.0%, P < 0.001).

Perceptions of Risk

Surgeons who were adequately vaccinated estimated that 0.5% (interquartile range, 0.1%–1.5%) of true percutaneous exposures to HBV-infected blood would result in their becoming acutely infected. Surgeons who were inadequately vaccinated estimated that 1.0% (interquartile range, 0.15%–5.0%) of percutaneous exposures to HBV-infected blood would result in acute infections. For inadequately vaccinated surgeons, the true risk is between 10% and 30%.^21,22 Using this range as the reference, we found that 82% of inadequately vaccinated surgeons underestimated their risk, 14% accurately estimated their risk, and 4% overestimated their risk.

Of the 267 surgeons who had performed at least one operation on patients known or suspected to be infected with HBV, 180 (67%) reported having sustained no percutaneous exposures; the remaining 87 surgeons reported sustaining a percutaneous exposure during 32.4% (95% CI, 23.7%–41.1%) of such operations.

DISCUSSION

The incomplete vaccination of transplant surgeons was last documented more than a decade ago,¹² prior to the implementation of requirements that all U.S. healthcare workers be offered vaccination. The present study suggests that, despite this mandate, vaccination of transplant surgeons remains incomplete.

An important question regarding surgeons’ incomplete vaccination is whether it is due to inadequate appreciation of their risk, beliefs that vaccination would not substantially alter this risk, or some other reason. We found that most surgeons thought they sustained no percutaneous exposures to blood while operating on HBV-infected patients. However, an estimated 6% of surgeries actually result in percutaneous exposures for surgeons even when universal precautions are strictly followed.²³ In addition, we found that surgeons who were inadequately vaccinated substantially underestimated the risk of patient-to-surgeon HBV transmission following an exposure.

By contrast, our finding that adequately vaccinated surgeons expressed less fear of becoming infected intraoperatively than inadequately vaccinated surgeons may suggest that surgeons are aware that vaccination reduces their risk. Thus, it seems that negligence of risk is more likely to explain surgeons’ incomplete vaccination than is negligence of vaccination benefit. Furthermore, the finding that inadequately vaccinated surgeons were fearful of acquiring hepatitis B despite underestimating the risk of transmission suggests that education regarding true transmission risks could increase fear even further, perhaps to a point where more surgeons would volunteer for vaccination.

This study is limited by its reliance on self-report. Evaluating surgeons’ reported vaccination, rather than actual vaccination, would likely result in our underestimating the proportion of inadequately vaccinated surgeons because the social desirability of preventive health behaviors²⁴ may lead respondents to overreport vaccination. A second limitation is that by only soliciting the total number of HBV vaccine injections surgeons had received, without considering the timing of these injections or whether the vaccine induced appropriate immunity, we may have misclassified some surgeons with regard to adequacy of vaccination. For example, surgeons who reported receiving 3 or more injections may still have been inadequately vaccinated if they had an inadequate immune response (defined as <10 IU/L antihepatitis B surface antigen). It is also true that some surgeons may be immune after receiving only 2 doses of vaccine, but they would be unlikely to know their immunity because antibody titers are typically not evaluated until after the third injection. Finally, because older surgeons were less likely to respond and less likely to be adequately vaccinated, nonresponse to the questionnaire may also lead to an underestimation of the proportion of inadequately vaccinated surgeons.

Our study suggests that at least 9 inadequately vaccinated transplant surgeons were exposed to HBV over a 2-year period without seeking appropriate postexposure evaluation. Both failure to be vaccinated against HBV and failure to be evaluated for active immunization after exposure reduce the safety of surgeons and their patients.

Physicians would seem free to decline interventions that would reduce their own risks of adverse outcomes. But we agree with previous commentators that physicians have a duty not to place their patients at undue risk for infection.^25,26 Given the availability of a safe and effective HBV vaccine, the substantial social investment in the training of surgeons, and the documented risks of surgeon-to-patient transmission of HBV,^27,28 it is hard to tolerate physicians’ choices to remain unvaccinated.

This paper does not address the question of what to do about surgeons, or other operating room personnel, who are already infected with HBV. However, we think that, if surgeons were better informed regarding the risks of intraoperative HBV transmission of HBV, they would all be vaccinated in an effort to protect both themselves and their patients. Because our study shows that a large proportion of surgeons underestimate this risk, we suggest that the most expedient way to protect patients and other surgical staff is to require documentation of appropriate HBV vaccination and detectable immunity for all surgeons and surgical staff.