• 
• 

Three Part Question

In [runners] does [not drinking fluids] prevent [exercise-related transient abdominal pain or stitch] compared to [drinking fluids]?

Clinical Scenario

A 23 year old women started a graduated training programme for 10 weeks since she set her goal to run 5 km nonstop. While running, she encounters stabbing pain under the lower edge of the ribcage that is inconvenient to continue her training. Therefore, she asks you if exercise-related transient abdominal pain (or stitch) is preventable e.g. by not drinking fluids before and during the run.

Search Strategy

We searched the following databases from date of inception to 15th of June 2012:
1. The Cochrane Library using the Wiley interface: (MeSH descriptor Sports explode all trees OR MeSH descriptor Exercise explode all trees) AND (MeSH descriptor Abdomen, Acute explode all trees OR “stitch” OR “side ache” OR “side cramp” OR “side pain” OR “exercise-related transient abdominal pain”)
2. MEDLINE via PubMed interface: (“sports”[MeSH] OR “exercise”[MeSH]) AND (“abdomen, acute”[MeSH] OR “stitch”[TW] OR “side ache”[TW] OR “side cramp”[TW] OR “side pain”[TW] OR “exercise-related transient abdominal pain”[TW] OR “ETAP”[TW])
3. Embase via Embase.com interface: (‘sport’/exp OR ‘exercise’/exp) AND (‘acute abdomen’/exp OR ‘stitch’ OR ‘side ache’ OR ‘side cramp’ OR ‘side pain’ OR ‘exercise-related transient abdominal pain’ OR ‘ETAP’)

Search Outcome

1. The Cochrane library: 0 Cochrane reviews and 2 clinical trials were found, both relevant to the clinical question.
2. MEDLINE: 41 studies were found, no additional study was selected.
3. Embase: 69 studies were found, no additional study was selected.

Altogether 112 references were found by searching The Cochrane Library, MEDLINE and Embase, of which 2 were relevant and of sufficient quality. In detail, we retained evidence from the highest level we found and those from the level below. No further papers were found by scanning the references of relevant papers. All relevant papers are summarised in the table below.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Plunkett BT 1999 New Zealand	10 healthy men (age: 21 ± 2 yr, height: 180 ± 5 cm, and weight: 81 ± 8 kg [mean ± SD]); All subjects consumed 200-300 mL of water at 1.5-2h before the testing exercise; Shortly before exercise, subjects were subjected to a body-mass-adjusted volume (14 mL/kg) of each of four fluids: water, commercially available energy drink Exceed (main energy constituent is glucose polymers), decarbonated Coca-Cola (main energy constituent is sucrose), and a hypertonic solution of the sugar lactulose or no fluid; Each exercise session consisted of five 5-min bouts of hard running on a treadmill, with 10-min rest between each bout; There were several days between each testing exercise.	Crossover trial (randomised according to Latin Square design) (level of evidence: 2b)	Mean intensity of stitch during first bout	Significantly more with each type of fluid compared to no fluid (p<0.0001)	Small study population; lack of objective method of verifying the intensity of stitch; no statistical analysis was performed to compare no fluid versus fluid consumption in second and third bout; allocation concealment is not described.
			Mean intensity of stitch by the fourth and fifth bout	Significantly more with each type of fluid (except for Exceed) compared to no fluid (p=0.003-0.02)
Morton et al, 2004, Australia	40 subjects (30 males and 10 females; age: 21.0 ± 0.5 yr, height : 177.1 ± 1.4 cm and weight: 71.9 ± 1.9 kg [mean ± SEM]; susceptible to exercise-related transient abdominal pain [ETAP]); All subjects consumed 500 mL of water at 2h before testing exercise; Shortly before and during exercise, subjects were subjected to a body-mass-adjusted volume (2 mL/kg at start and during exercise) of each of three fluids: flavored water, sports drink an reconstituted fruit juice or no fluid; Each exercise session consisted of 1 trial of 23 min running on a treadmill; There was approximately 1 week between each testing exercise.	Crossover trial (randomised according to Latin Square design)(level of evidence: 2b)	Incidence of ETAP (percentage of subjects)	No fluid: 40%;Flavoured water: 70% (p<0.05 compared to no fluid); Sports drink: 70% (p<0.05 compared to no fluid); Fruit juice: 83% (p<0.05 compared to no fluid)	Rate of fluid ingestion was high beyond that which occurs in a typical exercise situation and which is recommended by the Armerican College of Sports Medicine (600 to 1000 ml/h); lack of objective method of verifying the intensity of stitch; allocation concealment is not described).
			Severity of ETAP (mean ± SEM)	Mean severity: No fluid: 0.4 ± 0.1; Flavoured water: 0.6 ± 0.1 (p>0.05 compared to no fluid); Sports drink: 0.8 ± 0.2 (p>0.05 compared to no fluid); Fruit juice: 1.3 ± 0.2 (p<0.05 compared to no fluid)
			Duration of ETAP (min ± SEM)	No fluid: 6.8 ± 0.7 min; Flavoured water: 9.8 ± 1.2 min (p>0.05 compared to no fluid); Sports drink: 13.2 ± 1.4 min (p<0.05 compared to no fluid); Fruit juice: 14.4 ± 1.0 min (p<0.05 compared to no fluid)

Comment(s)

As all subjects in both studies consumed water before the testing exercise, stitch as a consequence of dehydration seems unlikely. All subjects were asked to fast for at least 5 h (Morton et?al) or to abstain from eating in the morning (Plunkett and Hopkins) before the trial, but not all test persons followed the ‘eating protocol’ (ie, 40% of the subjects of the trial of Plunkett and Hopkins preferred to consume their normal breakfasts). Only statistical tests (ie, p values) and conclusions about the comparison of fluid versus no fluid intake were described in the evidence table above and the clinical bottom line below, because the three part question deals with this comparison.

Editor Comment

ETAP, exercise-related transient abdominal pain

Clinical Bottom Line

In conclusion, drinking large quantities immediately before exercise is associated with exercise-related stitch.

References

1. Plunkett BT, Hopkins WG. Investigation of the side pain induced by running after fluid ingestion. Med Sci Ports Exerc 1999,31(8):1169-1175.
2. Morton DP, Aragón-Vargas LF, Callister R. Effect of ingested fluid composition on exercise-related transient abddominal pain. Int J Sport Nutr Exerc Metab 2004,14(2):197-208.