Article Text

Download PDFPDF

Less experience and running pace are potential risk factors for medical complications during a 56 km road running race: a prospective study in 26 354 race starters—SAFER study II
  1. Karen Schwabe1,
  2. Martin P Schwellnus1,2,
  3. Wayne Derman1,2,
  4. Sonja Swanevelder3,
  5. Esme Jordaan3,4
  1. 1Clinical Sport and Exercise Medicine Research Group, UCT/MRC Research Unit for Exercise Science and Sports Medicine, Department of Human Biology, Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa
  2. 2International Olympic Committee (IOC) Research Centre, University of Cape Town, Cape Town, South Africa
  3. 3Biostatistics Unit, Medical Research Council of South Africa, Parow, South Africa
  4. 4Department of Statistics and Population Studies, University of the Western Cape, Cape Town, South Africa
  1. Correspondence to Professor Martin P Schwellnus, Clinical Sport and Exercise Medicine Research Group, UCT/MRC Research Unit for Exercise Science and Sports Medicine, Department of Human Biology, Faculty of Health Sciences, University of Cape Town, South Africa, 3rd Floor, Sports Science Institute of South Africa, Boundary Road, Newlands, Cape Town 7700, South Africa; mschwell{at}iafrica.com

Statistics from Altmetric.com

Introduction

It has been documented that distance running,1 ,2 including ultra-marathon running,3–5 can be associated with medical complications in a variety of organ systems. In keeping with the recent interest in protecting the health of the athlete, it is important to identify risk factors for medical complications during ultra-marathon running so that medical care facilities can be planned for race day, and effective prevention programmes can be developed and tested.

Although the incidence of medical complications during marathon running1 and multistaged ultra-marathon running3 has been reported, the incidence and types of medical complications in single-day 56 km ultra-marathon running events have been reported only recently.5 In this study, we showed an overall incidence of 12.98 medical complications per 1000 runners starting a 56 km ultra-marathon race (1 in 77 runners starting the race), with an incidence of serious life-threatening medical complications of 0.65 (1 in 1538 runners starting the race). Furthermore, we showed that the most common medical complication in 56 km runners (per 1000 starters) was postural hypotension (2.77/1000), followed by serious exercise-associated muscle cramping (sEAMC; 1.9/1000), gastrointestinal complications (1.86/1000), musculoskeletal complications (1.82/1000) and dermatological complications (1.48/1000).5

There are very few published prospective cohort studies that report risk factors for medical complications in runners of ultra-marathons or similar endurance events. There are some studies that identified age,3 ,6 gender,3 ,7 running intensity (running pace),8 ,9 running experience10 and environmental conditions11 as possible risk factors for some specific medical complications that can occur during endurance events. However, we are not aware of any systematic investigation to determine independent risk factors associated with medical complications during single-day ultra-marathon running events. This information is vital to (1) design prevention programmes and thereby reduce the risk of medical complications (serious and less serious) during ultra-marathon running and (2) to plan optimal medical care during community-based endurance events.

One such event is the Two Oceans ultra-marathon (56 km), which takes place annually during late summer in Cape Town, South Africa. This popular and growing race draws over 6000 national and international entrants per year. The aim of this study was to identify independent risk factors that are associated with the development of medical complications during a single-day ultra-marathon running event. In addition, we wanted to determine independent risk factors associated with specific, more common medical conditions in different organ systems in ultra-marathon runners.

Methods

Type of study

This is a prospective study that consists of a 4-year cohort (2008–2011) of runner data collected through the race registry as well as medical admissions data.

Participants and demographics

This study forms part of a series of studies that were initiated to determine the incidence and risk factors for adverse medical events in endurance runners, and to then develop strategies to reduce this risk—the Strategies to reduce Adverse medical events For the ExerciseR (SAFER) studies.12 It was part of a large prospective cohort study that was conducted in all the participants of the Two Oceans Marathon races (including a 21 km half-marathon and a 56 km ultra-marathon) over a 4-year period (2008–2011).5 Details of the study methodology have been described.5 In summary, during the 4-year study period, a total of 26 354 runners started the 56 km races (85.4% of all registrations for the 56 km race) and only runners who started the race were included as participants in this study.

This cohort of 26 354 runners consisted of 19 998 men and 6356 women. Race entrants are required to be 20 years or older, and there is no upper limit to the age. The demographics by year of participation, age groups (<30, 31–40, 41–50 and >50 years) and gender of all the race starters in the 4-year study period is depicted in table 1.

Table 1

Demographics of all 56 km race starters (gender, age groups and year of participation)

Medical data collection and incidence of medical complications

Full details of the medical data collection procedure have been reported.5 In summary, accurate and comprehensive medical data concerning all medical complications were recorded during the 4-year study period. A ‘medical complication’ was defined as a medical condition that required medical care on race day that was of sufficient severity to warrant a medical assessment by a doctors, either in the medical facility at the end of the race, or at one of the referral hospitals (for runners who were assessed by medical staff on the route). Medical complication data entered into the database were recorded in a standardised format and included the organ system affected (cardiovascular, respiratory, fluid and electrolyte imbalance, central nervous system, thermoregulatory, musculoskeletal, gastrointestinal, dermatological, metabolic, renal and other) and the specific final diagnosis.

Incidence of medical complications

The crude incidence of any medical complications was calculated as medical admissions per 1000 runners who started the races, and this has been reported by year of observation and race type.5 These incidences were also reported in subgroups of runners by race type, system affected and final diagnosis.5 In this study, the incidence of medical complications will additionally be reported by previous running experience in the Two Oceans 56 km races (<1, 2–4, >5 previous medals) and running pace (<6, 6–7, >7 min/km).

Statistical analysis of data

All data were entered into an Excel spreadsheet (Microsoft 2010) and then analysed using the SAS (V.9.3) statistical program (SAS Institute Inc, Cary, North Carolina, USA). The medical complications data were analysed with a Poisson regression model, using a robust error estimator (log link function). This cohort consists of correlated data as 43% of runners ran this race more than once during the 4-year period. The correlated structure was accounted for using an unstructured correlation matrix. This was to estimate the incidence rates (IRs) and CI. Group comparisons and 95% CIs for these IRs and differences were also obtained. Regression analyses were conducted to determine risk factors associated with the development of any medical complication during 56 km running and then separately for the five most common medical complications that occurred. Risk factors investigated included year of race, gender, age group, running experience category and running pace category. The year of race was included in the analyses as a proxy for environmental conditions on race day to determine if it was a risk for medical complications. The mean (SD) temperature, humidity, rainfall, wind speed and calculated Wet Bulb Globe Temperature (WBGT) index during the race period (6:00 to 12:00 on race day) for each year of race have been reported previously.5

The five most common complications were postural hypotension, sEAMC, gastrointestinal complications, musculoskeletal complications and dermatological complications.5 Risk factors for these common medical complications were analysed. Risk factors were chosen on the basis that (1) there are reports that these factors are related to risk of medical complications,3 ,6–11 and that these data were recorded accurately in the study period.5 Risk factors therefore included the following: year of observation (as a proxy for environmental conditions), gender, age, running experience and running pace. The adjusted model for any medical complication was reported with all risk factors present irrespective of if they were significant, but for the five most common specific complications, the adjusted models only included the significant risk factors due to the small numbers present.

Results

Incidence and risk factors associated with the development of any medical complication during 56 km running

The incidence (12.98/1000 runners starting the race; 95% CI 11.67 to 14.43) of any medical complication during a 56 km race by year of observation (environmental conditions), gender, age group, running experience and running pace is depicted in table 2.

Table 2

The incidence (per 1000 runners starting the race: 95% CI) of any medical complication during a 56 km race by year of observation (environmental conditions), gender, age group, running experience group and running pace group

Univariate regression analysis

The crude unadjusted analysis showed that there was no significant difference in the incidence of any medical complication during a 56 km race by gender (p=0.7046), or age category (p=0.8130). However, there was a significant difference in the incidence by year of observation (environmental conditions) (p=0.0080), with a higher incidence in (1) 2009 (ambient temperature (mean±SD) of 17.1±1.4°C compared with both 2008 (ambient temperature of 18.2±1.9°C; p=0.0101) and 2010 (ambient temperature of 16.3±0.8°C; p=0.0034) and (2) 2011 (ambient temperature of 11.5±2.1°C) compared with 2010 (ambient temperature of 16.3±0.8°C;p=0.0270; table 3).

Table 3

Environmental conditions on race day for each year

There was also a significant difference in the incidence by running experience category (p=0.0061), with a higher incidence in runners with the least running experience category (≤1 medal) compared with runners in the intermediate category (2–4 medals; (p=0.0018) and runners in the most experienced category (≥5 medals; p=0.0183). Finally, there was a significant difference in the incidence by running pace category (p=0.0001). Runners in the fastest pace category (<6 min/km; p=0.0063) and runners in the slowest pace category (>7 min/km; p<0.0001) had a higher incidence of any medical complication compared with runners in the intermediate pace category (6–7 min/km).

Multiple regression analysis

In the adjusted model, factors associated with an increased risk of developing any medical complication in a 56 km race were: year of observation (environmental conditions; 2009 vs 2008, p=0.0176; 2009 vs 2010, p=0.0007; 2010 vs 2011, p=0.0112), less running experience (≤1 medal vs 2–4 medals, p=0.0097) and both fastest (<6 vs 6–7 min/km, p=0.0051) and slowest (>7 vs 6–7 min/km, p<0.0001) running pace categories.

Incidence and risk factors associated with the development of postural hypotension during 56 km running

In our study, the incidence of postural hypotension was 2.77/1000 runners starting the race (95% CI 2.20 to 3.48; 1 in every 361 race starters). The incidence (per 1000 runners starting the race) of postural hypotension during a 56 km race by year of observation (environmental conditions), gender, age group, running experience and running pace is depicted in table 4.

Table 4

The incidence (per 1000 runners starting the race: 95% CI) of postural hypotension during a 56 km race by year of observation (environmental conditions), gender, age group, running experience group and running pace group

Univariate regression analysis

The crude unadjusted analysis showed that there was no significant difference in the incidence of postural hypotension during a 56 km race by gender (p=0.3852), year of observation (environmental conditions) (p=0.5628), age category (p=0.3786) and category of running experience (p=0.6681). However, there was a significant difference in the incidence by running pace category (p=0.0260). Runners in the slowest category (>7 min/km) had a higher incidence of postural hypotension compared with runners in the intermediate pace category (6–7 min/km; p=0.0163) and runners in the fastest pace category (<6 min/km; p=0.0039).

Multiple regression analysis

In the adjusted model, the only factor associated with an increased risk of developing postural hypotension in a 56 km race was slow running pace (>7 vs 6–7 min/km, p=0.0163; >7 vs <6 min/km, p=0.0039).

Incidence and risk factors associated with the development of sEAMC during 56 km running

In this study, the incidence of sEAMC was 1.90/1000 runners starting the race (95% CI 1.44 to 2.50; 1 in every 526 race starters). The incidence (per 1000 runners starting the race) of sEAMC during a 56 km race by year of observation (environmental conditions), gender, age group, running experience and running pace is depicted in table 5.

Table 5

The incidence (per 1000 runners starting the race: 95% CI) of serious Exercise Associated Muscle Cramping (sEAMC) during a 56 km race by year of observation (environmental conditions), gender, age group, running experience group and running pace group

Univariate regression analysis

The crude unadjusted analysis showed that there was no significant difference in the incidence of sEAMC during a 56 km race by year of observation (environmental conditions; p=0.1686), and category of running experience (p=0.4325). However, there was a significant difference in the incidence by gender (p=0.0482), age category (p=0.0072) and running pace category (p=0.0231). Male runners had a higher incidence of sEAMC compared with female runners (p=0.0482). Runners in the 31–40 year category had a higher incidence of sEAMC compared with runners in the 41–50 year category (p=0.0393). Runners in the oldest age category (>50 years) also had a higher incidence of sEAMC compared with runners in the 41–50 year category (p=0.0007). Runners in the fastest pace category (<6 min/km) had a higher incidence of sEAMC compared with runners in the intermediate pace category (6–7 min/km; p=0.0066) but not compared to the runners in the slowest category (>7 min/km; p=0.0921).

Multiple regression analysis

In the adjusted model, the factors associated with an increased risk of developing sEAMC in a 56 km race were older age (>50 vs 20–30 years, p=0.0775; >50 vs 31–40 years, p=0.0219; >50 vs 41–50 years, p=0.0004) and faster running pace (<6 vs 6–7 min/km, p=0.0027; <6 vs >7 min/km, p=0.0339). When comparing the older age (>50 years) with the younger age (20–30 years), there was only a marginal significant difference (p=0.0775). It should be noted that there were only six runners with sEAMC in the young age group, and this limits this power of the analysis.

Incidence and risk factors associated with the development of gastrointestinal complications during 56 km running

In this study, the incidence of gastrointestinal complications was 1.86/1000 runners starting the race (95% CI 1.41 to 2.46; 1 in every 538 race starters). The incidence (per 1000 runners starting the race) of gastrointestinal complications during a 56 km race by year of observation (environmental conditions), gender, age group, running experience and running pace is depicted in table 6.

Table 6

The incidence (per 1000 runners starting the race: 95% CI) of gastrointestinal complications during a 56 km race by year of observation (environmental conditions), gender, age group, running experience group and running pace group

Univariate regression analysis

The crude unadjusted analysis showed that there was no significant difference in the incidence of gastrointestinal complications during a 56 km race by gender (p=0.9282), year of observation (environmental conditions; p=0.3962), age category (p=0.5871) and category of running experience (p=0.2654). However, there was a significant difference in the incidence by running pace category (p=0.0082). Runners in the slowest pace category (>7 min/km) had a higher incidence of gastrointestinal complications compared with runners in the intermediate pace category (6–7 min/km; p=0.0369) and runners in the fastest pace category (<6 min/km; p=0.0020).

Multiple regression analysis

In the adjusted model, the factors associated with a decreased risk of developing gastrointestinal complications in a 56 km race were slower running pace (>7 vs 6–7 min/km; p=0.0408: >7 vs <6 min/km; p=0.0019). Although there was an overall age group and gender interaction (p=0.0392), indicating that the risk of gastrointestinal complications in men and women was not necessarily the same within each age group, no individual differences in the incidences were significant.

Incidence and risk factors associated with the development of musculoskeletal complications during 56 km running

In this study, the incidence of musculoskeletal complications was 1.82/1000 runners starting the race (95% CI 1.37 to 2.42; 1 in every 549 race starters). The incidence (per 1000 runners starting the race) of musculoskeletal complications during a 56 km race by year of observation (environmental conditions), gender, age group, running experience and running pace is depicted in table 7.

Table 7

The incidence (per 1000 runners starting the race: 95% CI) of musculoskeletal complications during a 56 km race by year of observation (environmental conditions), gender, age group, running experience group and running pace group

Univariate regression analysis

The crude unadjusted analysis showed that there was no significant difference in the incidence of musculoskeletal complications during a 56 km race by gender (p=0.3463) and year of observation (environmental conditions; p=0.3315). The incidence of musculoskeletal complications was only marginally different by age category (p=0.0567) and running pace category (p=0.0736).

There was a significant difference in the incidence of musculoskeletal complications by category of running experience (p=0.0287). There was a higher incidence of musculoskeletal complications in runners with the least experience (≤1 medal) compared with runners in the intermediate category (2–4 medals; p=0.0341) and runners in the most experienced category (≥5 medals; p=0.0069).

Multiple regression analysis

In the adjusted model, none of the explored factors were associated with an increased risk of developing musculoskeletal complications in a 56 km race (p>0.05).

Incidence and risk factors associated with the development of dermatological complications during 56 km running

In this study, the incidence of dermatological complications was 1.48/1000 runners starting the race (95% CI 1.08 to 2.03; 1 in every 676 race starters). The incidence (per 1000 runners starting the race) of dermatological complications during a 56 km race by year of observation (environmental conditions), gender, age group, running experience and running pace is depicted in table 8.

Table 8

The incidence (per 1000 runners starting the race: 95% CI) of dermatological complications during a 56 km race by year of observation (environmental conditions), gender, age group, running experience group and running pace group

Univariate regression analysis

The crude unadjusted analysis showed that there was no significant difference in the incidence of dermatological complications during a 56 km race by gender (p=0.5700), year of observation (environmental conditions; p=0.2426) and age category (p=0.1613). In the incidence of dermatological complications, there was only a marginal difference between the categories of running experience (p=0.0750).

There was a significant difference in the incidence of dermatological complications by running pace category (p=0.0130). Runners in the fastest pace category (<6 min/km) had a higher incidence of dermatological complications compared with runners in the intermediate pace category (6–7 min/km; p=0.0179) and runners in the slowest pace category (>7 min/km; p=0.0166).

Multiple regression analysis

In the adjusted model, the only factor associated with an increased risk of dermatological complications in a 56 km race was a faster running pace (<6 vs 6–7 min/km, p=0.0179; <6 vs >7 min/km; p=0.0166).

Summary of risk factors associated with the development of medical complications during 56 km running

Independent risk factors associated with any medical complication, and the most common specific medical complications (postural hypotension, sEAMC, gastrointestinal, musculoskeletal and dermatological) in a 56 km race, are depicted in table 9.

Table 9

A summary of risk factors for any medical complications and more common medical complications in the multiple model during a 56 km race

Discussion

In this study, the incidence and risk factors for any medical complication, and specific common medical complications during a 56 km race, are reported. A regression model was used to determine independent risk factors that may be associated with the development of (1) any medical complications and (2) more specific common medical complications during 56 km running. Risk factors that were entered into the model were: year of observation (a proxy for environmental conditions), gender, age, running experience and running pace.

The main findings of this study were, first, that running pace (both slowest and fastest pace categories), less running experience and years of observation (environmental conditions) were risk factors for the development of any medical complication during a 56 km race. Second, risk factors associated with specific common medical complications were identified as follows: postural hypotension (slow running pace), sEAMC (older age; fast running pace), gastrointestinal complications (slow running pace) and dermatological complications (fast running pace).

Running pace, specifically running in the slowest running pace category, was the most important risk factor, as this category had the highest risk of developing any medical complication and, more specifically, postural hypotension or gastrointestinal complications. We are aware of only one study where running pace was investigated as a possible risk factor for medical complications in runners participating in a 7-day multistaged ultra-marathon race,3 and in this study, slower running pace was not related to an increased risk for medical complications. However, this is a different race type, and running pace as a risk factor for medical complications during a single-day ultra-marathon has not been reported. An obvious possible explanation for our observation is that medical complications that developed during the race resulted in a slower punning pace. However, as we do not have accurate data on the running pace variation during the race, the time of onset of the medical complication during the race and prerace medical complications or disease, we cannot exclude other possible reasons for our observation. Although our model did take into account age and gender, we did not have any data on other variables that may also determine slow running pace, particularly training history. Therefore, although a slower running pace was associated with the risk for medical complications, including postural hypotension and gastrointestinal complications, the cause-effect relationship of this factor would have to be explored in future studies.

In our study, a faster running pace was predictive of any medical complications and, indeed, specific complications such as sEAMC and dermatological complications. The association between running at a faster running pace and EAMC has been postulated previously13 and documented in some studies in runners8 and triathletes.14 The possible mechanisms for this have also been explored and may be related to the development of premature fatigue resulting in abnormal neuromuscular control.15 The results of this study therefore support the hypothesis that running at a faster pace is a risk factor for the development of sEAMC.

We are aware of only one study where risk factors for dermatological complications in ultra-marathon runners competing in a 7-day staged event have been reported.3 In that study, no specific risk factors for dermatological complications were identified. In our study, most of the dermatological complications were skin abrasions, friction injuries and cuts from minor falls (data not reported), and this is similar to that reported in marathon runners1 and ultra-marathon runners.3 Although speculative, a faster running pace may be associated with an increased risk of friction injuries and minor falls. The Two Oceans races attract large fields and there is overcrowding, particularly at the start of the race. Runners can be advised to use antichafing creams and exercise caution to reduce the risk of falling when running at a faster pace. Finally, we did not have any data on prerace dermatological conditions in these runners, and this may have to be explored in future studies.

In our study, age was not a risk factor for medical complications in general and other common specific medical complications. The exception was that older age was associated with an increased risk of sEAMC. Our data are in support of the results reported in one of the earliest cross-sectional studies.16 However, older age was not shown to be a risk factor associated with EAMC in another prospective study in runners with EAMC8 or in case–control studies in endurance athletes with a history of EAMC.14 ,17 Therefore, the association between older age and EAMC requires further study.

A novel finding in our study was that less running experience was a risk factor for any medical complications. As aforementioned, most of the dermatological complications were skin abrasions, friction injuries and cuts from minor falls. These skin conditions are preventable, and less running experience could well be associated with an increased risk of these complications. Runner education, targeted at the less experienced runner could reduce the risk of these complications.

Finally, the year of observation was predictive of any medical complication, but not any of the more common specific medical complications. We included this variable in the model because of the potential variation in environmental conditions that may occur from year to year. However, our data show that, in general, environmental conditions on race day (ambient temperature, humidity, rainfall, wind speed and WBGT index) were quite similar in the 4-year study period (table 1). The highest incidence of medical complications was recorded in 2009 and 2011. Environmental conditions in 2009 did not differ substantially from those in 2008 and 2010 (table 3), but the ambient temperature was lower in 2011. Therefore, aside from colder conditions in 2011, there is no obvious pattern to explain the effect of the year of observation through variation in environmental conditions. Factors other than environmental conditions, such as the risk profile (other than age, gender, race pace, race experience) of the race starters, may have played a role in the higher incidences recorded in 2009 and 2011. These would have to be explored in future studies.

The strengths of this study are that it is, to the best of our knowledge, the largest prospective study until now to determine possible risk factors for any medical complication and specific common medical complications during a 56 km ultra-marathon race. Furthermore, we collected comprehensive and accurate data on all runners who registered and started the races, and the medical records for all medical complications. A recognised limitation of our study was that we could not include all the possible intrinsic and extrinsic risk factors for medical complications in our model as these data were not available. In future studies, additional prerace medical and training history, as well as history of medical complications, would have to be included in models that assess risk factors for medical complications in ultra-marathon runners.

In summary, our study identified, for the first time, risk factors for all medical complications and specific common medical complications in runners participating in a 56 km race. These data can have important practical clinical implications in that ultra-marathon runners who are at higher risk of developing any medical complication and, indeed, some more common specific medical complications can now be identified. For example, older runners can now be advised that running at a faster pace may increase their risk of sEAMC. Similarly, race directors can identify the inexperienced runners as a group with a higher risk of medical complications and plan medical care appropriately. Furthermore, these data form the basis of further studies to identify risk factors for medical complications in endurance runners and then develop prevention strategies to reduce complications.

What are the new findings?

  • To the best of our knowledge, the Strategies to reduce Adverse medical events For the ExerciseR (SAFER) II study is the first study to systematically document risk factors for all medical complications and specific common medical complications during running in a population of ultra-marathon (56 km) runners.

  • Running pace (both the slowest and fastest pace categories), less running experience and the years of observation (environmental conditions) were potential risk factors for the development of any medical complication during a 56 km race.

  • Independent risk factors associated with specific common medical complications were as follows: postural hypotension (slow running pace), serious exercise-associated muscle cramping (older age; fast running pace), gastrointestinal complications (slow running pace) and dermatological complications (fast running pace).

How might it impact on clinical practice in the near future?

  • Clinicians taking care of distance runners can now start to identify subgroups of ultra-marathon runners who may be at higher risk of developing any medical complication and specific common medical complications during races.

  • Data from this study will form the basis for further clinical studies to develop and then test the effects of intervention strategies to reduce the risk of adverse medical events during ultra-marathon running.

Acknowledgments

The authors would like to acknowledge all the medical staff for their service to the athletes participating in the Two Oceans Marathons over the years, the race organisers for their support of the project, and all the athletes for their interest and participation in the races.

References

View Abstract

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.