Send letter to journal:
Re: Authors reply

Dear Editor

Re: eLetter from Adriana Suely de Oliveira Melo et al.

Authors responses are in bold:

Barakat et al. [1] have presented us with a paper of excellent methodological quality, following all the steps recommended in the Consolidated Standards of Reporting Trials (CONSORT) and dealing with a question that never fails to generate controversy with respect to the practice of physical activity during pregnancy: prematurity. Another strongpoint of the paper is the fact that the physical exercise was systematized and monitored, guaranteeing that the pregnant woman indeed followed the prescribed program.

Various controversies continue to surround the topic of physical exercise and pregnancy and the real effects of exercise on the conceptus remain to be clarified. The spectrum of these effects ranges from fetal growth to the duration of the pregnancy, with some studies associating prematurity and growth restriction with the practice of physical exercise [2-4]. Despite these speculations, until recently no randomized clinical trials (RCT) with adequate sample sizes had been identified in which pregnant women were systematically followed up for a period encompassing the second and third trimesters.

Authors Response
We acknowledge the authors of the letter for their nice comments. Please see our responses and comments below.

The excellent quality of this paper prompted us to examine it in detail in an attempt to understand some points that we would now like to put to the authors. Since the objective of the RCT was to evaluate the risk of premature labor, would it not have been better to have excluded all the pregnant women with a history of premature labor in view of the fact that the results show that one of the cases of prematurity in the intervention group was precisely due to a prior history of prematurity?

Authors Response
History of premature labour is not an absolute contraindication for exercising during pregnancy. It is actually a relative contraindication. In fact, we only studied women with no history of recurrent spontaneous preterm birth, i.e., number of previous preterm deliveries ≤1. Anyway, the potential confounding factor arising from including women with one prior preterm delivery was controlled for as the % of women with such history was very similar (and actually very low) in both groups (please see Table 1). As indicated in the paper, of the two women of the training group showing preterm delivery, only one (gestational age: 36wk 2d) had previous history of preterm delivery (n=1). We judged interesting to analyse individually what happened with gestational outcome in those cases of one previous preterm labour. This could only be achieved by studying women with prior history of prematurity.

Another point that drew our attention concerns the exclusions in both groups, which were the result of various situations that may have affected the outcome "gestational age", such as bleeding, pregnancy- induced hypertension and threatened preterm labor. In our opinion, these women should have continued in the study and an intent-to-treat analysis should have been carried out. We were also intrigued by the fact that one patient was excluded because her pregnancy was a twin pregnancy. Was a single pregnancy not one of the inclusion criteria?

Authors Response
Indeed, it would have been interesting to study individually some 'case reports' with such conditions (same as we did with women with one prior preterm delivery). Bleeding, pregnancy-induced hypertension and threatened preterm labor are however absolute contraindications for exercise during pregnancy. As such, there were considered to be exclusion criteria. The ACOG guidelines establish that exercise during pregnancy must be safe. As for twin pregnancy, the first inclusion criteria for study participants (as indicated in the paper) was being "i) gravida with singleton and uncomplicated gestation" (i.e., women with twin pregnancy were excluded from our study). We believe that future RCTs are needed including solely women with twin pregnancy.

It may perhaps have been interesting NOT to have included women With a history of premature delivery. Although the inclusion criteria Accepted the possibility of the participants having had at the most one previous premature delivery, this may have had an effect on the mean gestational age reported in the present study.

Authors Response
Please see our response above, which shows that this was not a confounding factor at all.

We were unable to identify in the paper any description of the parameters used to calculate sample size to determine whether the final number of participants included was sufficient to demonstrate any differences between the groups. Could a type II statistical error have occurred?

Authors Response
We employed a conservative approach for sample size estimation. Power calculations were made for the primary outcome measures of gestational age, Apgar score, birth weight and length. It was determined that adequate power (>0.80) would be achieved with 72 pregnant women in the training group and with 70 pregnant women in the non-exercising control group. All power computations assumed that comparisons of baseline to 26-week* scores would be tested at the 5% significance level. All power computations allowed for 10% dropouts over 26 weeks. *(The duration of the training program was of 26 weeks).

Another minor question we would like to pose is whether the intensity of the prescribed exercise was light-to-moderate or moderate, since it is described in different ways in the various sections of the manuscript and it is known that some outcomes are dependent on the intensity of exercise.

Authors Response
Training intensity was ≤ light-to-moderate (< moderate). Indeed, heart rate was consistently ≤80% age-predicted maximum HR value (220 minus age) and we included toning and very light resistance exercises.

Finally, we would like to know whether the authors have data on Other gestational or perinatal outcomes, since such a well-conducted RCT as This one should have generated interesting results that deserve to be published.

Authors Response
Yes, we do have a complete set of data (included post-natal follow-up), some of which will be analysed in further studies. As for the present paper, we did not judge it adequate to include such a big amount of data as many of them are not really related to mean gestational outcome per se and it would deviate readers¡¯ attention from the main study findings.

References

1. Barakat R, Stirling JR, Lucia A. Does exercise training during pregnancy affect gestational age? A randomised controlled trial. Br J Sports Med 2008; 42(8):674-8.

2. De Ver Dye T, Fernandez ID, Rains A, Fershteyn Z. Recent studies in the epidemiologic assessment of physical activity, fetal growth, and preterm delivery: a narrative review. Clin Obstet Gynecol 2003; 46(2):415- 22.

3. Grisso JA, Main DM, Chiu G, Synder ES, Holmes JH. Effects of physical activity and life-style factors on uterine contraction frequency. Am J Perinatol 1992; 9(5-6):489-92.

4. Misra DP, Strobino DM, Stashinko EE, Nagey DA, Nanda J. Effects of physical activity on preterm birth. Am J Epidemiol 1998; 147(7):628-35.

Send letter to journal:
Re: Does exercise training during pregnancy affect gestational age? A randomised controlled trial

Dear Editor

Barakat et al. [1] have presented us with a paper of excellent methodological quality, following all the steps recommended in the Consolidated Standards of Reporting Trials (CONSORT) and dealing with a question that never fails to generate controversy with respect to the practice of physical activity during pregnancy: prematurity. Another strongpoint of the paper is the fact that the physical exercise was systematized and monitored, guaranteeing that the pregnant woman indeed followed the prescribed program.

Various controversies continue to surround the topic of physical exercise and pregnancy and the real effects of exercise on the conceptus remain to be clarified. The spectrum of these effects ranges from fetal growth to the duration of the pregnancy, with some studies associating prematurity and growth restriction with the practice of physical exercise [2-4]. Despite these speculations, until recently no randomized clinical trials (RCT) with adequate sample sizes had been identified in which pregnant women were systematically followed up for a period encompassing the second and third trimesters.

The excellent quality of this paper prompted us to examine it in detail in an attempt to understand some points that we would now like to put to the authors. Since the objective of the RCT was to evaluate the risk of premature labor, would it not have been better to have excluded all the pregnant women with a history of premature labor in view of the fact that the results show that one of the cases of prematurity in the intervention group was precisely due to a prior history of prematurity? Another point that drew our attention concerns the exclusions in both groups, which were the result of various situations that may have affected the outcome “gestational age”, such as bleeding, pregnancy-induced hypertension and threatened preterm labor. In our opinion, these women should have continued in the study and an intent-to-treat analysis should have been carried out. We were also intrigued by the fact that one patient was excluded because her pregnancy was a twin pregnancy. Was a single pregnancy not one of the inclusion criteria?

It may perhaps have been interesting NOT to have included women with a history of premature delivery. Although the inclusion criteria accepted the possibility of the participants having had at the most one previous premature delivery, this may have had an effect on the mean gestational age reported in the present study.

We were unable to identify in the paper any description of the parameters used to calculate sample size to determine whether the final number of participants included was sufficient to demonstrate any differences between the groups. Could a type II statistical error have occurred?

Another minor question we would like to pose is whether the intensity of the prescribed exercise was light-to-moderate or moderate, since it is described in different ways in the various sections of the manuscript and it is known that some outcomes are dependent on the intensity of exercise.

Finally, we would like to know whether the authors have data on other gestational or perinatal outcomes, since such a well-conducted RCT as this one should have generated interesting results that deserve to be published.

References

1. Barakat R, Stirling JR, Lucia A. Does exercise training during pregnancy affect gestational age? A randomised controlled trial. Br J Sports Med 2008; 42(8):674-8.

2. De Ver Dye T, Fernandez ID, Rains A, Fershteyn Z. Recent studies in the epidemiologic assessment of physical activity, fetal growth, and preterm delivery: a narrative review. Clin Obstet Gynecol 2003; 46(2):415- 22.

3. Grisso JA, Main DM, Chiu G, Synder ES, Holmes JH. Effects of physical activity and life-style factors on uterine contraction frequency. Am J Perinatol 1992; 9(5-6):489-92.

4. Misra DP, Strobino DM, Stashinko EE, Nagey DA, Nanda J. Effects of physical activity on preterm birth. Am J Epidemiol 1998; 147(7):628-35.