Polycythemia Prevalence and Risk Factors in Pilots
INTRODUCTION: Pilots are frequently exposed to thrombotic risk as a result of immobility from air travel. As hypoxemia is associated with secondary polycythemia, and polycythemia increases the risk of thrombosis, intermittent exposure to high-altitude hypoxic environments could escalate the risk of thrombosis in pilots. Our objectives were to find the prevalence of polycythemia in airplane pilots (primary outcome) and to assess associated risk factors of polycythemia (secondary outcome). METHODS: This study is a cross-sectional descriptive study. Data was collected from paper-based and computerized medical records of airplane pilots who applied for Class 1 Aviation Medical Certificate renewal at the Institute of Aviation Medicine, Royal Thai Air Force, Bangkok, Thailand, in 2018. The data was sampled by a simple random sampling technique. RESULTS: A total of 386 paper-based records were sampled. Of those, 29 (7.5%) of the pilots met polycythemia criteria. Spearman’s correlation analysis showed a significant positive correlation between Body Mass Index (BMI) and hemoglobin (correlation coefficient = 0.127) and between BMI and hematocrit (correlation coefficient = 0.105). In multivariate logistic regression of each variable on polycythemia as defined by hemoglobin alone, piloting a non-pressurized aircraft was found to be an independent predictor of polycythemia (odds ratio = 4.3). DISCUSSION: The prevalence of polycythemia in airplane pilots was 7.5%. Operating a non-pressurized aircraft was a significant risk factor of polycythemia, and pilots with higher BMI were more likely to have increased red blood cell parameters. Thanapaisan P, Plaingam M, Manyanont S. Polycythemia prevalence and risk factors in pilots. Aerosp Med Hum Perform. 2024; 95(9):683–687.
Polycythemia is a state in which there are excessive red blood cells (RBC) in circulation. Etiologies of polycythemia can be categorized into primary and secondary causes. Primary polycythemia, known as polycythemia vera, is a disease of abnormally increased erythropoiesis by the bone marrow as a result of a genetic defect. Secondary polycythemia is mediated by erythropoietin (EPO) secretion. High altitude, smoking, lung diseases, and cardiovascular diseases are common causes of hypoxemia leading to EPO secretion and, consequently, secondary polycythemia. Intermittent hypoxic exposure is also associated with secondary polycythemia.12,19,21
The airplane cockpit shares a ventilation and pressurization system with the rest of the cabin. When cruising, modern airplanes’ altitude typically starts from 1813 ft (553 m),7 with a mean pressurized cabin altitude of 6309 ft (1923 m) for airplanes with a pressurization system.15 The Federal Aviation Administration (FAA) regulates a ceiling of 12,500 ft (3810 m) for unpressurized flights without oxygen supplement, above which the flight cannot be more than 30 min.6 The cabin altitude has a negative effect on inspired partial pressure of oxygen, resulting in a decrease of alveolar oxygen partial pressure.
= alveolar partial pressure of oxygen
= inspired gas fraction of oxygen
= atmospheric pressure
= saturated water vapor pressure at alveolar temperature
= alveolar partial pressure of carbon dioxide
= respiratory gas exchange ratio
From the alveolar gas equation, at the altitude of 6309 ft (1923 m), the alveolar partial pressure of oxygen will be reduced to 69.62 mmHg in account of a normal saturated vapor pressure of 47 mmHg, a normal alveolar carbon dioxide partial pressure of 40 mmHg, and a 0.8 normal respiratory gas exchange ratio. For non-pressurized aircraft pilots, the degree of alveolar hypoxia is even more severe should they operate above 6309 ft (1923 m) without oxygen supplement. Because of the lower alveolar partial pressure of oxygen, the arterial partial pressure of oxygen will consequently decline, leading to a decrease in hemoglobin oxygen saturation and, ultimately, a reduction in oxygen delivery. Specialized interstitial cells in renal cortex will respond by increasing EPO production in an attempt to reconstitute the oxygen delivery. However, this has an undesirable side-effect of elevating RBC concentration. Therefore, airplane pilots could be at risk of secondary polycythemia, since they are exposed to such hypoxic environments intermittently. Considering that pilots are already frequently exposed to thrombotic risk as a result of immobility from air travel longer than 3–8 h,9,11,20 polycythemia could further escalate the risk.
Our study’s primary objective is to determine the prevalence of polycythemia in airplane pilots. Assessing associated risk factors of polycythemia is the secondary objective.
METHODS
Subjects
This study is a cross-sectional study. Data was collected from medical records of airplane pilots who applied for Class 1 Aviation Medical Certificate renewal at the Institute of Aviation Medicine, Royal Thai Air Force (RTAF), Bangkok, Thailand, between January 1, 2018, and December 31, 2018. The study was consistent with the Declaration of Helsinki and was approved by the ethics committee of Bhumibol Adulyadej Hospital Research Center (IRB No.20/66).
Procedure
Airplane pilots were sampled by simple random sampling from paper-based medical records at the Institute’s Administrative Department. Each pilot’s data was collected from their Application for Medical Certificate Form CAAT-AMG-502, issued by The Civil Aviation Authority of Thailand in April 2017, filled and submitted to the Institute of Aviation Medicine, RTAF, by pilots and their respective medical examiners throughout the year 2018. The data of each one’s single latest visit in 2018 was gathered. The process continued until the data of 386 samples were accumulated. With regard to the insufficiency of information from paper-based medical records, each of the samples were matched to the computerized medical record database and the remaining information of each sample was completed. Exclusion was not applied to analyze the prevalence of the polycythemic population. For further analysis, however, pilots who met any of these criteria were excluded: medication use (specifically steroid, diuretic, iron, vitamin B12, and folate), diagnosed heart disease, diagnosed lung disease, diagnosed hematological disease (specifically thalassemia trait, thalassemia disease, iron deficiency anemia, leukemia, lymphoma, polycythemia vera, or secondary polycythemia), pregnancy, menstrual abnormality, obstructive sleep apnea (OSA), unspecified illness, or unspecified medication.
Polycythemia was defined as hemoglobin > 16.5 g ⋅ dL−1 or hematocrit > 49 in men and hemoglobin > 16.0 g ⋅ dL−1 or hematocrit > 48 in women, in accordance with the World Health Organization (WHO) 2016 criteria for polycythemia vera.2 Established evidence has demonstrated that these thresholds are associated with a higher risk of major adverse cardiovascular events (MACE) irrespective of fulfilling WHO polycythemia vera criteria,10 and the definition was widely accepted as a diagnostic cut point for polycythemia in various hematological studies.13,22 Flight time per week was calculated by flight time in the last six months (hr), divided by 26 wk. Aircraft type was classified by cockpit pressurization. A319, A320, A321, A330, A350, A380, B737, B747, B767, B777, B787, Q400, ATR72, C750, SAAB340, and B350 are pressurized aircraft, while DA42, DA40, C152, C172, C182, C208, PC6, and CASA.C-212 are non-pressurized aircraft. Living altitude was estimated from the mean altitude at the province in each participant’s address. Smoking status was clustered into “smoker” and “nonsmoker.” Hypertension was defined as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, or currently using antihypertensive medications.
Statistical Analysis
Data was analyzed by R for Windows, version 4.3.1. Continuous data was reported as mean ± SD. If the data did not distribute normally, it would be reported as medians and interquartile ranges. Categorical data was reported as percentages and frequencies. Correlations between continuous variables and hemoglobin and hematocrit were depicted with Pearson or Spearman’s correlation coefficients, depending on the normality. Differences of means or medians of continuous variables between polycythemia and non-polycythemia groups were assessed using Student’s t-test. Mann-Whitney U-test was applied to compare data that was heteroscedastic or not in the normal distribution. Frequencies of categorical variables of the two groups were analyzed with the Chi-square test or Fisher’s exact test when more than 20% of cells had expected frequencies less than five. Binary logistic regressions were utilized to establish whether polycythemia was independently associated with total flight time, flight time per week, aircraft type, and other potential variables. A two-tailed P-value < 0.05 was considered statistically significant.
RESULTS
From the medical record department at the Institute of Aviation Medicine, RTAF, 386 paper-based records were randomly sampled. Of those, 29 (7.5%) of the pilots met polycythemia criteria by either hemoglobin or hematocrit definition, as presented in Table I. The samples’ nationality was all Thai. Five of the pilots who reported either unspecified illness, unspecified medication, or OSA, met exclusion criteria and were excluded from further analysis. Scale variables were tested by the Kolmogorov-Smirnov test for normality, but none were normally distributed; therefore, Mann-Whitney U-test and Spearman’s correlation coefficients were applied. Missing data was treated as absent in each variable separately. The remaining were considered valid samples. Between polycythemia and non-polycythemia groups, none of their characteristics had significant differences, except for hemoglobin and hematocrit, which were categorizing variables.
In correlation analysis, samples missing the data of any analyzed variables were excluded only from those variables’ analyses. Between hemoglobin and hematocrit, Spearman’s coefficient of 0.88 (P < 0.001) suggests a strong positive relationship. As portrayed in Table II, the Spearman’s correlation analysis also showed a significant positive correlation between BMI and hemoglobin (correlation coefficient = 0.127, P = 0.013). A slightly less but also significant correlation occurred between BMI and hematocrit (0.105, P = 0.040). There were negative correlations among living altitude, total flight time, and flight time per week against both hematological parameters, though none were statistically significant.
In the multivariate logistic regression of each variable on polycythemia, gender and hypertension were not included due to insufficient comparison data. The total number of analyzed samples was reduced to 282 after the omission of samples that were missing at least one variable data. Hosmer and Lemeshow Test showed no significant expected values from the analysis. The multivariate logistic regression showed that there were no variables that showed significant association to polycythemia, considering that it incorporated both the hemoglobin and hematocrit definition for polycythemia. Given that polycythemia was defined by hemoglobin alone (Table III), piloting non-pressurized aircrafts was found to be an independent predictor of polycythemia, with an OR of 4.3 and 95% confident interval of 1.051–16.222, P = 0.034. Living altitude was observed to have a weak association with polycythemia. However, the two-tailed P-value was 0.05, rendering the OR significance unclear and exacerbating its evident lack of clinical significance. On the contrary, multivariate logistic regression analysis in terms of hematocrit-based polycythemia showed that no variables were associated with its occurrence, as illustrated in Table III.
DISCUSSION
Upon reviewing the literature, the authors believe that this study might be the first to describe polycythemia status in pilots. The primary objective was satisfied, determining that the prevalence was 7.5% of the airplane pilot population in Thailand. Further study of the prevalence of polycythemia in the general Thai population is required so that the result of this study can be compared.
The association between aircraft pressurization and polycythemia has been demonstrated in our study, and the method of multivariate logistic regression indicates that the relationship is independent. Considering that pilots tend to fly at the highest possible altitude for speed and fuel efficiency, the degree of alveolar partial pressure of oxygen could be as worrisome as 47 mmHg, if flying at the FAA exact recommended limit of 12,500 ft (3810 m) in an unpressurized airplane without oxygen supplement. Our finding was consistent with this concept, and it confirmed that piloting an unpressurized aircraft could cause hypoxia to an extent that RBC production is stimulated. Advising the use of oxygen masks, even when flying below 12,500 ft (3810 m), might be suggested for non-pressurized airplane pilots in order to ensure their circulating oxygen sufficiency. However, despite the fact that the association was statistically significant, the 95% confidence interval for the OR was wide, implying substantial variability in the data. Moreover, the scarcity of descriptive information, such as each operation’s altitude, supplemental oxygen usage, flight duration, and ground periods between flights, was a limitation of our study and posed as a possible confounder in the relationship between aircraft pressurization and polycythemia.
Flight time, both lifelong and weekly, failed to demonstrate their consequences to hematological parameters in our study. Their extremely broad interquartile ranges, as depicted in Table I, indicated that there was a high variability of flight time among the pilots and skewed distributions. These characteristics influenced the results of our study along with the effect of intermittent hypoxia. RBC production in response to intermittent hypoxia is dynamic, depending on factors including the severity of hypoxia, the duration of each exposure, and the normoxic period after each exposure. Unfortunately, most of these data points were either insufficiently recorded or not recorded at all in the present study, highlighting an area that warrants further investigation.
From the correlation analysis, BMI showed a weak positive relationship with both hemoglobin and hematocrit. Although statistically significant, a correlation at a very low level (<0.5) might imply any of the following: 1) the practical importance of this relationship could be negligible, 2) limited predictive power, or 3) the possibility of nonlinear relationships. However, it is widely recognized that BMI serves as a significant risk factor for OSA.3 As OSA is linked to higher RBC parameters,12,16 an increase in BMI could lead to an elevation in RBC concentration, aligning with the findings from our research. An alternative explanation is that there is a positive connection between BMI and excessive sweating.8 The rise in hemoglobin and hematocrit among individuals with a high BMI could be linked to increased evaporative sweat loss and a reduction in plasma volume. A high BMI could potentially bring about many adverse outcomes. Hence, it is crucial that practitioners emphasize to pilots the importance of maintaining a normal BMI and stress the significance of adequate hydration for individuals with higher BMI.
Although previous studies demonstrated the positive interrelation between living altitude, smoking status, and polycythemia,4,5,14 in our study, the results were not consistent with those reports. The effects of living altitude and cigarette smoking appeared insignificant in our study. This might presumably be due to the lack of descriptive details, such as the duration of exposure to both altitude and smoking. In addition, we reported smoking status as smoker and nonsmoker due to the paucity of data. Previous research has shown that the risk of polycythemia corresponds with the intensity of smoking,1,17,18 meanwhile, in our collected data, the intensity of smoking was indiscriminate. It is possible that the rarity of heavy smokers in pilot population was the factor that misled the analysis. The same could be assumed for high-altitude residents in our study, which were evidently few.
In the present study, we did not compare our data to general population norms in Thailand due to the absence of a clinically acknowledged upper normal limit for RBC parameters that Thai practitioners widely accept, apart from the WHO criteria for polycythemia vera. Potential socio-cultural factors, such as genetic predispositions and nutritional intake, may influence hemoglobin and hematocrit levels. While we attempted to control for these potential confounders through random sampling, the availability of a well-defined normal range of RBC indices in the Thai population, provided by further studies, would certainly aid in demonstrating the impact of flight on polycythemia status more definitively.
Finally, in our report, we intentionally chose data from 2018 to ensure that the data was from active pilots. The pandemic of the COVID-19 infection had a devastating consequence on the airline industry starting from the year of 2019, halting numerous airplane operations. The event has barred us from conducting an observational study on the effect of flights on human pathophysiology ever since.
In summary, the observed prevalence of polycythemia in the airplane pilot population in Thailand was 7.5%. Non-pressurized aircraft increased the risk of polycythemia, and BMI showed a positive relationship with both hemoglobin and hematocrit. Future recommendations should include supplementing oxygen in non-pressurized flights below 12,500 ft (3810 m) and stressing the importance of maintaining a normal BMI.
Contributor Notes
