Abstract
Statement of Problem Since the 1968 Summer Olympic Games (Mexico City, altitude ~ 2200m above sea level (SL)), a numerous amount of research has been conducted on the effects of hypoxic manipulation (HM) (i.e. altitude training which includes LHTH, LHTL, LLTH, and IHE) for improving SL exercise performance. However, the SL performance results of HM studies are inconclusive. Further, the results of the physiological variables such as VO2max, hematocrit (Hct), and hemoglobin concentration [Hb] are conflicting. Therefore, it is unclear the effects HM may have on SL exercise performance and certain physiological variables. A method of resolving the conflicting data on HM studies is the use of meta-analysis. Therefore, the purpose of this study was to conduct a meta-analysis on HM studies to identify the effects of HM versus normoxic training (NT) on SL exercise performance, VO2max, Hct, and [Hb] in trained athletes. Methods A literature search from 1965 to 2008 was conducted to locate pertinent studies on HM. An inclusion criteria was developed to determine studies that were included in the analysis. Studies included in the analysis met the following inclusion criteria: 1) adequate use of a control group (CG), 2) published performance, or VO2max, or Hct, or [Hb] results with means ± SD and 3) use of trained athletes. If studies met the inclusion criteria, they were read and coded for the following variables: 1) HM model which included LHTH, LHTL, LLTH, and IHE, 2) performance were defined as time in a time trial, or peak power output during a GXT, or total work capacity 3) VO2max as well as Hct and [Hb]. To calculate effect size (ES) the following formula was used: Cohen’s d = (Postm – Prem)/PreSD. ES were than corrected (ESCorr) for sample biasness and then weighted mean of the ES (ESWM) was calculated. Twenty-nine studies met the inclusion criteria for performance with a total of 42 and 49 ESCorr extracted from NT and HM, respectively. Twenty-seven studies met the inclusion criteria for VO2max with a total of 39 ESCorr extracted from NT and HM. Eighteen studies met the inclusion criteria for [Hb] with 23 ESCorr extracted from NT and HM. For Hct, 14 studies met the inclusion criteria with 15 and 17 ESCorr extracted from NT and HM, respectively. Results are reported in means ± 95% CI. Results For performance, HM (0.27 ± 0.02) was statistically greater than NT (0.17 ±0.02). The only two HM models that were statistically greater than NT were LHTH and LLTH. The ESWM ± 95% CI for LHTH and LLTH were 0.38 ± 0.27 and 0.37 ± 0.09, respectively. For VO2max, the only HM model that was statistically greater than NT was LLTH (0.37 ± 0.12). For [Hb], the only HM model that was statistically greater than NT was LHTH (0.53 ± 0.27). There were no significant differences between groups for Hct. As a whole, in the HM group, there was a significant correlation between the ESCorr for VO2max and performance (r = 0.67, p = 0.00008), however no significant correlation was observed between HM ESCorr for [Hb] or Hct and VO2max. Conclusion In conclusion, as a whole, HM is more beneficial to improving SL performance when compared to SL training alone. In addition, the greatest benefits from HM are from LHTH and LLTH. Further, the results of VO2max, [Hb], and Hct between LHTH and LLTH are unclear. Therefore, it is still uncertain what factor(s) mediate the improved SL performance.