Implementation of Smoke-free Policy in University Hospital Decreases Carboxyhemoglobin Level in Inpatients Undergoing Surgery

Over the past two decades, many hospitals in advanced countries have declared a “smoke-free hospital,” and adoption of such a policy has recently, albeit very belatedly, begun in Japan. Because cigarette smoking is a risk factor for mortality and morbidity, this implementation may be expected to produce measurable benefits for the majority of patients, provided it effectively reduces smoking in the hospital environment.1In cigarette smokers, as well as in passive smokers, blood carboxyhemoglobin concentration (COHb) is known to be elevated.2Herein we report a significant reduction in COHb values in surgical inpatients following the implementation of such a policy.

We compared COHb values before and after the implementation of a smoke-free hospital (on April 1, 2003) after more than a year’s preparation. We collected all arterial blood oxymetery data (measured using an ABL700, Radiometer, Copenhagen, Denmark) obtained from those inpatients undergoing surgery who had an arterial puncture or an arterial line in place for collection of arterial blood samples in the operating room. Arterial blood samples taken just before or after the induction of anesthesia were immediately subjected to the measurement of arterial blood gas tensions and COHb. The implementation of a smoke-free university campus was begun on April 1, 2005. Differences in COHb were examined via  a one-way analysis of variance with an unpaired t  test (with a Bonferroni correction) being used for post hoc  comparisons.

As shown in figure 1, the mean values COHb were 1.65 ± 0.87% (n = 656, mean ± SD) over the 3 months before the implementation of a smoke-free hospital and 1.15 ± 0.50% (n = 614) just after the implementation, and this decreased COHb level remained stable. After the implementation of a smoke-free university campus (April 2005), it showed a slight decrease to 0.98 ± 0.40% (n = 713) over the next 3 months. There was no difference in age distribution, hemoglobin concentration, or arterial oxygen tension (Paco²) before and after the implementation. Whereas in 2002 the percentage of surgical patients who were smokers was 26.9% and the average hospital stay before surgery was approximately 6.7 days, in 2005, these were smaller (22.5% and 5.4 days, respectively). Among outpatients, the mean COHb values were 1.74 ± 0.94% (n = 1,069) for 12 months before the implementation of the smoke-free hospital and 1.64 ± 0.72% (n = 1,475) after the implementation of a smoke-free university campus.

These data document that the implementation of a smoke-free hospital caused a dramatic decrease in COHb values among surgical inpatients. It is unlikely that upon implementing a smoke-free hospital, all patients who smoked had stopped before admission. However, such patients were no longer able to smoke in the hospital buildings and had an approximately 6-day period of forced abstinence or reduction of smoking before surgery after the policy implementation. These data could also preclude the possibility that seasonal variations in COHb among the population admitted to our hospital might have affected the results, as COHb did not significantly change during the observation period, except the timing of the implementation of the smoke-free hospital. The implementation of a smoke-free policy may have made smokers abstain from smoking or prevented inpatients from passive smoke. Thus, such a policy seems to be an effective way of reducing smoke pollution in the hospital environment and lowers the carbon monoxide concentration in the blood of inpatients.

Although COHb may not be the sole factor in determining smoke-induced morbidity and mortality, smokers are at higher risk of cardiopulmonary3and wound-related postoperative complications than nonsmokers,4and high COHb levels are associated with high mortality.5Thus, this first observation of a decrease in hospital-wide COHb warrants further study to evaluate the potential benefits to patients resulting from admission to a smoke-free hospital, as well as to people in a smoke-free community.

The authors thank Mr. H. Jintoku (Radiometer, Copenhagen, Denmark) for his help collecting part of the data.

*Gifu University Graduate School of Medicine, Gifu, Japan. shu-dohi@gifu-u.ac.jp