Safety of Vojta’s reflex locomotion in term pregnancy: a randomized controlled pilot study of uterine activity and labor onset

Abstract

The Vojta’s method (VRL) is a neurophysiological rehabilitation method used to support and induce reflex responses of locomotor and vegetative systems. It uses involuntary motor reaction of the body during pressure stimulation of so-called trigger zones. Pregnancy is currently considered a contraindication for the use of VRL due to the potential risk of inducing regular uterine activity and, consequently, triggering labor. The aim of the study is to evaluate changes in uterine activity during Vojta’s reflex locomotion and to determine whether the method is associated with the induction of labor. Secondary goal is to assess its safety in term pregnancy and its potential use for rehabilitative purposes. This single-center, single-blinded, parallel-group randomized controlled pilot trial included 40 pregnant patients between 40 + 0 to 41 + 0 weeks of gestational age. Participants were randomized (sealed envelope method) to receive either stimulation of trigger zones according to VRL (n = 20) or sham stimulation (n = 20). Cardiotocographic (CTG) recordings were conducted right before and immediately after stimulation and evaluated by two independent obstetricians. Additionally, participants completed a questionnaire evaluating subjective responses to stimulation. Primary outcomes were CTG-detected uterine changes and time from stimulation to delivery. Secondary outcomes included self-reported sensations and pain intensity (VAS). None of the patients delivered as a result of VRL stimulation. CTG showed increased uterine activity in 45% of VRL stimulated participants vs. 10% in the control group. Median time to labor onset in the VRL group was 5 days (1–8), compared to 4 days (1–9) in the control group. VRL was well tolerated (mean VAS 1.85), but rated significantly less pleasant than sham (p = 0.043). Our findings suggest that the use of VRL in term pregnancy is likely safe and well tolerated. While uterine activity was observed in 45% of cases, this activity was not sufficient to induce labor. Our findings, however, do not support pregnancy as a blanket contraindication.

Trial registration: This clinical trial was registered at http://clinicaltrials.gov (ID: NCT06339528).

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-31871-9.

Introduction

Reflex locomotion according to Vojta’s therapy is a neurorehabilitative concept originally developed for the treatment of infants at risk of non-physiological psychomotor development, especially those with spastic cerebral palsy. This concept utilizes the stimulation of specific zones described by the founder of Vojta’s reflex locomotion (VRL) – Václav Vojta – to evoke comprehensive motor and non-motor responses¹.

Trigger zones are specific areas on the body that are stimulated with pressure during VRL. According to Václav Vojta, there are 10 identified trigger zones on the human body. These proprioceptive zones are stimulated through pressure to induce reflexive responses. Vojta’s method relies on the spatial and temporal summation of signals from these zones. The variation in the combination of stimulated zones (spatial summation) enhances the multiplicative effect of afferent impulses. Sufficiently prolonged stimulation of trigger zones (temporal summation) leads to isometric contractions in specific muscle groups as described by Vojta. The motor response to trigger zone stimulation is involuntary and manifests through symmetrical and asymmetrical movements of the limbs, trunk, neck, and head. These movements are organized into two primary motor patterns: reflex creeping and reflex rolling¹.

Several clinical indications support the use of VRL. In children, key indications include central developmental disorders, cerebral palsy, coordination disorders, and scoliosis. In adults, VRL is applied in the treatment of various conditions affecting the musculoskeletal system, such as vertebrogenic disorders, post-stroke rehabilitation, muscular disorders, and others^2–4.

When planning treatment using VRL, it is crucial to consider contraindications. Major contraindications include acute inflammation or infection, severe cardiovascular disease, acute joint or muscle pain or injury, severe neurological conditions, acute respiratory issues, and pregnancy due to the potential risk of inducing uterine contractions. Stimulation during pregnancy is described as contraindicated; however, some studies are emerging that aim to objectively assess changes in the human body following VRL stimulation^5–7. This pilot study aims to investigate the potential of VRL in initiating uterine activity and explore its feasibility as a new pre/induction method. To this date there are two primary methods of inducing labor - pharmacological and non-pharmacological. The most commonly used non-pharmacological methods include cervical ripening using a balloon catheter with membrane sweeping is efficient in 20% of patients⁸ or amniotomy, which can reduce duration of labor by approximately 2 h⁹. Pharmacological methods of induction include administration of prostaglandins via vaginal or oral routes¹⁰.

Materials and methods

Study design and participants

Prospective, single-center, single-blinded, parallel-group randomized controlled pilot study was carried out in the University Hospital of the Third Faculty of Medicine, Charles University in the Department of Gynecology and Obstetrics. The inclusion period was from October 2021 through December 2022. Patients qualified if they were at term pregnancy (between 40 + 0 and 41 + 0 weeks of gestational age), had a singleton pregnancy, with a Bishop/cervix score ≥ 6 and did not experience serious pregnancy complications (such as preeclampsia, gestational hypertension, intrahepatic cholestasis of pregnancy or high-risk gestational diabetes). Term pregnancies were selected due to the lack of prior scientific evidence on the effects of VRL stimulation and to avoid potential risks to a preterm fetus. The Bishop score threshold (≥ 6) was chosen to ensure cervical readiness for vaginal delivery, as inducing contractions with an unprepared cervix could potentially lead to obstetric complications. Exclusion of pregnancies with complications aimed to minimize confounding factors that might influence the physiological response to stimulation. Patient’s tolerance of the supine position was also necessary. Patients were enrolled regardless of their gravidity and parity. A total of 40 patients were eligible for inclusion in the study, randomized, and allocated into two groups of 20 participants each. Randomization was conducted using the sealed envelope method.

Ethics statement

The study was approved by the Ethics Committee of the Faculty Hospital Kralovske Vinohrady (EK-VPl421012021) and retrospectively registered on ClinicalTrials.gov (ID: NCT06339528; registration date: 01/04/2024), URL: https://clinicaltrials.gov/study/NCT06339528. All participants provided written informed consent prior to inclusion in the study. All procedures were performed in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments.

Patients characteristics

40 pregnant patients participated in this study. The study population consisted of White (Caucasian) women of Czech origin. Parity and gravidity data are summarized in Tables 1 and 2. None of the multiparous women had a previous cesarean delivery (no TOLAC cases). Information on comorbidities and gestational history is also provided in Tables 1 and 2. They were equally divided into two groups of 20 participants each: a study group and a control group. Participant flow through the study is presented in the CONSORT diagram (Supplementary Fig. 1).

Table 1.

Study group - patient characteristics.

Patient	Age	Gestational age at use of Vojta’s method	Gravida	Para	Comorbidities	Gestational age at delivery	Time from stimulation to delivery (days)	Mode	Mode of delivery	Effect of Vojta´s method
1	32	40 + 2	1	0	NO	41 + 1	6	Induction	Vaginal	NO
2	29	40 + 5	2	0	NO	41 + 3	5	Induction	Vaginal	NO
3	30	41 + 0	1	0	NO	41 + 5	5	Induction	Vaginal	NO
4	35	40 + 3	3	1	NO	41 + 1	7	PROM	Vaginal	NO
5	31	40 + 5	1	0	NO	41 + 1	3	PROM	Vaginal	NO
6	23	40 + 3	1	0	Pregnancy related anemia	41 + 4	8	Induction	Vaginal	NO
7	26	40 + 3	2	1	NO	40 + 5	2	PROM	Vaginal	NO
8	37	40 + 3	3	1	History of DVT	41 + 4	8	Induction	Vaginal	NO
9	34	40 + 5	2	1	NO	40 + 6	1	Spontaneous contractions	Vaginal	NO
10	40	40 + 3	3	1	History of DVT	41 + 2	6	Spontaneous contractions	Vaginal	NO
11	31	40 + 5	1	0	Low risk gestational diabetes	41 + 1	3	Spontaneous contractions	Acute C-section	NO
12	32	40 + 1	1	0	NO	40 + 5	4	Spontaneous contractions	Vaginal	NO
13	31	40 + 5	1	0	Pregnancy related anemia	41 + 5	7	Induction	Vaginal	NO
14	34	40 + 3	2	0	Pregnancy related anemia	41 + 2	6	Spontaneous contractions	Vaginal	NO
15	36	40 + 1	2	1	Pregnancy related anemia	40 + 6	5	Spontaneous contractions	Vaginal	NO
16	30	40 + 1	1	0	Pregnancy related anemia	40 + 5	4	Spontaneous contractions	Vaginal	NO
17	37	40 + 0	3	2	NO	40 + 4	4	Spontaneous contractions	Vaginal	NO
18	32	40 + 5	1	0	NO	41 + 4	6	Induction	Vaginal	NO
19	33	40 + 1	1	0	NO	40 + 2	1	Spontaneous contractions	Vaginal	NO
20	31	40 + 3	4	2	NO	40 + 6	3	Spontaneous contractions	Vaginal	NO

Baseline and clinical characteristics of 20 pregnant participants who underwent Vojta’s reflex locomotion (VRL) stimulation between 40 + 0 and 41 + 0 weeks of gestation. Gestational ages are shown as completed weeks + days. Time from stimulation to delivery is expressed in days. PROM = premature rupture of membranes. DVT = deep vein thrombosis. C-section = cesarean section. Effect of Vojta’s method denotes whether uterine contractions following stimulation were considered to represent a potential clinical effect leading to labor onset (as evaluated by CTG and clinical outcome).

PROM – premature rupture of membranes; DVT – deep vein thrombosis; CTG – cardiotocography; VRL – Vojta reflex locomotion.

Table 2.

Control group - patient characteristics.

Patient	Age	Gestational age at use of Vojta’s method	Gravida	Para	Comorbidities	Gestational age at delivery	Time from sham stimulation to delivery (days)	Mode	Mode of delivery
1	30	40 + 2	2	0	Gestational anemia	40 + 3	1	Spontaneous contractions	Vaginal
2	37	40 + 6	2	1	NO	41 + 0	1	Spontaneous contractions	Vaginal
3	34	40 + 4	3	1	NO	41 + 3	8	Induction	Vaginal
4	37	40 + 4	2	1	Gestational anemia	41 + 0	3	Spontaneous contractions	Vaginal
5	29	40 + 5	3	1	NO	41 + 2	4	Induction	Vaginal
6	22	40 + 1	1	0	Gestational anemia	41 + 0	6	Spontaneous contractions	Vaginal
7	27	40 + 1	1	0	Hypothyreosis, pituitary microadenoma	41 + 3	9	Spontaneous contractions	Acute C-section
8	31	40 + 2	2	1	NO	41 + 0	5	Spontaneous contractions	Vaginal
9	28	40 + 3	1	0	NO	41 + 1	5	Spontaneous contractions	Vaginal
10	27	40 + 1	1	0	Low risk gestational diabetes, gestational anemia	40 + 5	4	Spontaneous contractions	Vaginal
11	24	40 + 0	2	1	Low risk gestational diabetes, gestational anemia	40 + 3	3	Spontaneous contractions	Vaginal
12	29	40 + 6	1	0	NO	41 + 1	2	Spontaneous contractions	Vaginal
13	23	40 + 0	2	0	NO	41 + 2	9	Spontaneous contractions	Acute C-Section
14	35	40 + 4	1	0	NO	41 + 4	7	Induction	Vaginal
15	30	40 + 1	2	1	NO	40 + 2	1	Induction	Vaginal
16	37	40 + 0	3	2	NO	40 + 4	4	Induction	Vaginal
17	29	40 + 5	2	0	Gestational anemia	40 + 6	1	Spontaneous contractions	Vaginal
18	32	40 + 0	3	2	Gestational anemia	41 + 0	7	Spontaneous contractions	Vaginal
19	33	40 + 2	2	0	NO	40 + 3	1	Spontaneous contractions	Acute C-Section
20	31	40 + 4	4	0	Hypothyreosis	41 + 3	1	Spontaneous contractions	Vaginal

Baseline and clinical characteristics of 20 pregnant participants who underwent sham stimulation between 40 + 0 and 41 + 0 weeks of gestation. The control intervention was applied at non-VRL zones. Gestational ages are presented as completed weeks + days. Time from sham stimulation to delivery is expressed in days. Mode of delivery indicates the final method of childbirth.

VRL – Vojta reflex locomotion; C-section – cesarean section.

Study group (details in Table 1.)

This group included 20 pregnant patients with singleton pregnancies. No serious pregnancy complications were observed. Five patients were diagnosed with mild pregnancy-related anemia, which was managed with oral substitution therapy. Two patients had a history of DVT (deep venous thrombosis) and one patient developed low-risk gestational diabetes during pregnancy. The median age of the patients was 32 years (range 23–40). 12 patients were nulliparous and eight were multiparous (six patients had previously given birth once, and two patients had given birth twice). All patients underwent stimulation according to VRL in between 40 + 0 and 41 + 0 weeks of gestational age. In the case of 13 patients, labor activity (either regular contractions or rupture of membranes) occurred by the latest 41 + 3 weeks of pregnancy (according to our guidelines - the latest day before start of preinduction or induction of delivery). In seven patients, spontaneous labor had not occurred by gestational age 41 + 3 and these patients were admitted for labor induction. 19 patients delivered vaginally and in one case acute C-section was performed due to obstetrical complication (risk of fetal hypoxia).

Control group (details in Table 2.)

This group also included 20 pregnant patients with physiological singleton pregnancies. Five patients were diagnosed with mild pregnancy-related anemia with per os ferrotherapy. Two patients were being treated for pre-existing hypothyroidism, and two patients developed low-risk gestational diabetes during pregnancy. The median age of the patients was 30 years (range 22–37). 11 patients were nulliparous and nine were multiparous (seven patients had previously given birth once, and two patients had given birth twice). All patients underwent sham stimulation in between 40 + 0 and 41 + 0 weeks of gestational age. In the case of 15 patients, labor activity occurred latest by 41 + 3 weeks of pregnancy and five patients were admitted for labor induction due to post-term pregnancy. 17 patients delivered vaginally and in three cases there was a need for acute C-section for the risk of fetal hypoxia).

Methods

Depending on group allocation, participants either underwent stimulation based on Vojta’s method (study group) or received a sham intervention (control group).

• Study group: Participants in this group received stimulation of trigger zones according to VRL, performed by licensed physiotherapists. The baseline position for both groups was “reflex rolling one,” where the lower limbs were semi-flexed at the knee joints to reduce tension on the lumbar spine. Figure 1 illustrates the thoracic trigger zone - the main stimulation site used in the study (intercostal space between the sixth and seventh ribs). Each patient was stimulated for 15 min on the left side and then for 15 min on the right side, according to standard clinical practice in Vojta reflex locomotion.

Fig. 1.

Trigger zone - the breast zone used for VRL stimulation. Illustration of trigger zone used in the study group. The physiotherapist applies proprioceptive pressure to the thoracic zone—intercostal space between the sixth and seventh ribs—while the patient is positioned in “reflex rolling one,” with lower limbs semi-flexed to minimize lumbar tension. This standardized positioning and contact were used in all VRL interventions. VRL – Vojta reflex locomotion.

• Control group: Participants in this group received sham stimulation performed by physicians at zones outside the designated pressure zones. Specifically, stimulation was applied four to five centimeters above the patella for 15 min and on the lateral malleolus for another 15 min, inspired by studies by Sanz-Esteban and Hok^11–13.

Data collection

Cardiotocographic (CTG) recordings were conducted right before and immediately after stimulation. Standard parameters such as basal rate, presence of accelerations and decelerations and variability of recording were evaluated in the cardiography records according to the FIGO guidelines¹⁴. The tocographic record evaluated the presence, intensity, and regularity of uterine activity. A contraction was defined as a transient increase in uterine tone with a rounded peak lasting ≥ 30 s. The number of contractions was quantified as the average number of uterine contractions per 10 min. Normal uterine activity was defined as ≤ 5 contractions per 10 min. The amplitude of contractions assessed on the tocogram was also evaluated, with an increase in the mean intensity by more than 10 mmHg considered a change. For the purpose of this study, change in CTG was operationalized as the appearance of new contractions where none were present before stimulation, or an increase in contraction frequency by ≥ 2 per 10 min compared with the pre-stimulation recording or an increase in the amplitude of pre-existing contractions by more than 10 mmHg.

Subsequently, the CTG recordings were independently reviewed by two obstetricians who were blinded to the treatment allocation. Neither assessor was present during the stimulation procedures, and both received anonymized CTG tracings for comparative evaluation. The CTG recordings were anonymized and coded before evaluation to ensure blinding of the assessors. Each record was labeled with a neutral identification code, and the sequence of “before” and “after” recordings was randomized. As a result, assessors were unable to infer group allocation or the order of recording. No unblinding occurred during or after the evaluation process. The tocographic traces were assessed with particular attention to the presence and frequency of uterine contractions, comparing findings before and after stimulation.

Additionally, all participants completed a structured questionnaire as presented in Supplementary Fig. 2. The questionnaire was administered immediately after the intervention and assessed the following information:

• Perceived uterine activity, assessed by a binary yes/no question asking whether participants noticed significant changes during stimulation of pressure zones.

• Character of uterine activity changes, collected via an open-ended question asking participants to describe the nature of the perceived changes if applicable.

• Pain during stimulation, assessed by a binary yes/no question asking whether participants experienced any pain during stimulation of pressure zones.

• Pain characteristics, described through an open-ended response detailing the type, location, and quality of pain, and rated on a numeric scale from 0 (“no pain”) to 10 (“unbearable pain”).

• Change in fetal activity, assessed by a binary yes/no question asking whether participants noticed any alteration in the baby’s movements during stimulation.

• Other physiological sensations, assessed by a binary yes/no question targeting changes such as deeper breathing or awareness of heartbeat during stimulation.

• Character of other bodily sensations, collected via an open-ended question describing the nature of such changes, if reported.

• Pleasantness of the stimulation, rated using a Visual Analogue Scale (VAS), where a score of 1 indicated “very pleasant” and a score of 10 indicated “very unpleasant”.

In addition to the core parameters described above, we also recorded the time interval between the intervention - VRL stimulation or sham stimulation - and delivery for all participants. This value was calculated as the number of days from the date of the intervention to the date of birth. This parameter was included to explore potential associations between the stimulation and the timing of labor onset.

Data regarding deliveries were collected and patients were verbally asked whether they felt that delivery was directly associated with the stimulation.

Data analysis

Descriptive analysis of the obtained data was carried out. To assess the association between the type of intervention (VRL stimulation vs. sham stimulation) and the subjectively perceived change in uterine activity, pain during stimulation, changes in fetal activity, and other bodily changes, the Fisher’s exact test (two-tailed) was used due to the small sample size and low expected frequencies in some cells (Table 3). To examine the correlation between group assignment and subjective discomfort during intervention, as measured by the Visual Analogue Scale (VAS), the Spearman rank-order correlation coefficient was calculated.

Table 3.

Comparison of subjective perception and CTG evaluation during Vojta therapy and Sham stimulation.

		Group
		Control group		Treatment group
		Count	Column N %	Count	Column N %	Value	Exact Significance (2-sided)
Subjective change in uterine activity	No	17	85,0%	14	70,0%	1,29	0.451
	Yes	3	15,0%	6	30,0%
Pain during stimulation	No	18	90,0%	15	75,0%	1.558a	0.407
	Yes	2	10,0%	5	25,0%
Change in child activity	No	10	50,0%	11	55,0%	.100a	1.000
	Yes	10	50,0%	9	45,0%
Other changes	No	17	85,0%	6	30,0%	12.379a	0.001
	Yes	3	15,0%	14	70,0%
Change in CTG (1st doctor)	No	18	90,0%	10	50,0%	7.619a	0.014
	Yes	2	10,0%	10	50,0%
Change in CTG (2nd doctor)	No	15	75,0%	11	55,0%	1.758a	0.320
	Yes	5	25,0%	9	45,0%
Change in CTG (at least one doctor)	No	14	70,0%	9	45,0%	2.558a	0.200
	Yes	6	30,0%	11	55,0%

Comparison of subjective and objective outcomes between participants undergoing Vojta reflex locomotion (VRL) stimulation and those receiving sham stimulation. The table summarizes categorical data for perceived uterine activity, pain, changes in fetal activity, and other bodily sensations, as well as CTG-evaluated uterine. Statistical analysis was performed using Fisher’s exact test (two-tailed).

Values: Count represents the number of participants within each response category; Column N (%) denotes the proportion within each group. Value = test statistic. Exact significance (2-sided) = corresponding p-value.

CTG – cardiotocography; VRL – Vojta reflex locomotion.

In both the control and intervention groups, the variable “time from stimulation to delivery (in days)” was analyzed. Because this variable showed a non-normal distribution, it was summarized using medians and interquartile ranges (IQR). Between-group differences were described using these non-parametric summary measures rather than parametric confidence intervals, as this approach is more appropriate for skewed data in a small pilot sample. The median time to delivery was calculated for each group in order to assess potential differences in labor onset timing (Table 4).

Table 4.

Comparison of time from stimulation to delivery (in days) between study and control groups.

Statistic	Study group (VRL)	Control group (sham)
Mean (SD)	4.7 (2.11)	4.1 (2.52)
Median (IQR)	5.0 (3.0–6.0)	4.0 (1.0–6.5)
Minimum – Maximum	1–8	1–9
Sample size (n)	20	20

Descriptive comparison of the interval (in days) from intervention to delivery between participants undergoing Vojta reflex locomotion (VRL) and those receiving sham stimulation. Continuous data are presented as median with interquartile range (IQR), mean ± standard deviation (SD), and range (minimum–maximum).

VRL – Vojta reflex locomotion; SD – standard deviation; IQR – interquartile range.

Inter-rater reliability between the two independent obstetricians evaluating the CTG recordings was assessed using Cohen’s kappa (κ) coefficient. The level of agreement was interpreted according to Landis and Koch (1977)¹⁵, where κ values of 0.61–0.80 indicate substantial agreement and values above 0.80 indicate almost perfect agreement. A 95% confidence interval (CI) was calculated for κ to quantify the precision of the estimate.

Safety assessment

Safety was assessed by performing the stimulation in the obstetric unit of the Department of Gynecology and Obstetrics under direct supervision of an obstetrician. A normal (physiological) CTG recording obtained immediately prior to stimulation was required as a precondition for participation. During each stimulation, a full obstetric team was present and prepared to perform an emergency caesarean section if required. No extended in-hospital observation period was conducted, as all participants were at term and had provided informed consent acknowledging the possibility of labour onset. After post-stimulation CTG and vaginal examination, patients were managed according to the standard outpatient protocol used after membrane sweeping, discharged home, and invited for follow-up in the antenatal clinic within 1–3 days.

Long-term maternal and neonatal outcomes, including Apgar scores, NICU admission, and postpartum complications, were not systematically collected, as the study was designed to assess only the immediate uterine and maternal response to stimulation. No deliveries occurred within 24 h of VRL or sham stimulation, and no acute adverse events were observed during or immediately after the intervention. Three participants later underwent caesarean delivery (two for failure to progress and one for intrapartum maternal fever with non-reassuring fetal status), but the time interval between stimulation and birth did not allow any causal relationship to be inferred.

Immediate safety outcomes were prespecified as:

• Tachysystole: > 5 uterine contractions per 10 min.

• Non-reassuring FHR: CTG patterns meeting FIGO non-reassuring criteria (e.g., reduced variability, recurrent late or prolonged decelerations, baseline abnormalities).

• Need for tocolysis: administration of any uterine relaxant prompted by excessive uterine activity or fetal status following the intervention.

These endpoints were assessed on paired CTG recordings obtained immediately before and after the intervention.

Results

Study group

No patients in the VRL group delivered within 24 h of stimulation. Two patients (9 and 19) delivered within 48 h after stimulation. Patient 9 presented to our clinic 26 h after VRL stimulation with regular uterine contractions, which led to cervical dilatation. Uterine activity commenced 24 h post-stimulation. Spontaneous rupture of membranes occurred 28 h after stimulation, and the patient delivered a healthy newborn 30 h post-stimulation. Patient 19 came to our clinic with regular uterine contractions 30 h after VRL stimulation, exhibiting an advanced cervical dilatation. Regular uterine contractions began approximately 27 h post-stimulation, and the patient delivered a healthy newborn 33 h after stimulation.

The median time to delivery after stimulation in the VRL group was 5 days (IQR 3–6; mean 4.7 ± 2.1 days) (Table 4). In the VRL stimulated group, uterine activity changed in terms of increase or start in uterine activity in nine cases (45%) in CTG records (Table 3). Figure 2 illustrates a representative example of tocographic activity before and after stimulation, demonstrating the transition from absent uterine activity to regular contractions. Inter-rater reliability between the two independent obstetricians evaluating the CTG recordings was substantial, with Cohen’s κ = 0.68 (95% CI [0.46–0.90], p < 0.001). This indicates a high level of agreement in the classification of uterine activity changes before and after stimulation¹⁵. Among these nine patients, contractions were mild to moderate in intensity, with tocographic amplitudes typically between 20 and 35 mmHg, occurring at irregular intervals of 6–10 min, and resolving spontaneously within 20–40 min after stimulation. No patient met the FIGO criteria for tachysystole (≥ 5 contractions in 10 min), and fetal heart rate patterns remained within normal limits throughout. None of these findings were associated with cervical changes or clinical signs of labor onset. These data indicate that although VRL stimulation transiently increased uterine activity, it had no clinically significant effects on maternal or fetal outcomes. In the questionnaire responses, six VRL stimulated patients (30%) reported subjective changes in terms of contractions. Also, according to the questionnaire, pain characterized as menstrual pain was described in five VRL stimulated patients (25%). Change in fetal movement was noted in nine cases of VRL stimulated patients (45%). Although patients stimulated with VRL generally reported low levels of discomfort, with an average VAS score of 1.85 ± 0.75 (SD) (where 1 represents “very pleasant” and 10 “very unpleasant”), statistical analysis revealed a significant positive correlation between group assignment and reported discomfort (Spearman’s ρ = 0.326, 95% CI [0.002–0.588], p = 0.043), suggesting that participants in the study group reported slightly higher discomfort compared to the control group.

Fig. 2.

Tocography before (A) vs. After (B) The stimulation according to Vojta therapy. Panel A shows no uterine activity during a 20-minute tocographic recording obtained prior to stimulation. Panel B demonstrates regular uterine contractions with amplitudes ranging between 40 and 55 mmHg following stimulation.

The statistical analysis demonstrated a highly significant association (p = 0.001) between the type of stimulation and responses to the “other changes” category in the questionnaire. These additional subjective and objective sensations - such as deepened breathing, sweating, reflexive hand movements, and feelings of heat - were significantly more frequently reported in the study group than in the control group, indicating a clear effect specific to the VRL intervention. They occurred in 12 VRL (60%) stimulated patients according to the questionnaire.

Control group

The median time to delivery after sham stimulation was 4 days (IQR 1–6.5; mean 4.1 ± 2.5 days) in the control group. The between-group difference in medians was 1 day, with markedly overlapping interquartile ranges (Table 4). In the control group, contractions occurred in two patients (10%) and 4 (20%) patients had decreased uterine activity in CTG records. According to the questionnaire, three patients (15%) experienced change in uterine activity. Also, according to the questionnaire pain characterized as menstrual pain was described in two patients (10%) in the control group. Change in fetal movement was noted in ten cases (50%) in the control group (Table 3). Sensations, such as deepened breathing, sweating, reflex rolling of hands and heat, occurred in three (15%) cases in the control group according to the questionnaire.

Discussion

In this study, two patients in the VRL group delivered within 36 h after stimulation (as detailed in the Results section) compared to six patients in the control group who also delivered within 36 h following sham stimulation. These findings do not indicate any association between VRL and the onset of labor. Also, there was no evidence of shortened time to delivery in the VRL group, where the median time to labor onset was 5 days (IQR 3–6; mean 4.7 ± 2.1 days) compared with 4 days (IQR 1–6.5; mean 4.1 ± 2.5 days) in the control group. This supports the safety of the method in term pregnancy.

To objectively assess the effect of VRL on uterine activity, cardiotocographic (CTG) monitoring was performed in all participants. In the VRL group, uterine contractions were observed in 9 patients (45%), whereas in the control group, contractions were present in 2 patients (10%). These objective findings are consistent with patients’ subjective reports collected via post-intervention questionnaires. A subjective perception of increased uterine activity was reported by 30% of patients in the VRL group, compared to 15% in the control group. Although VRL does not induce labor, it may be accompanied by a subjective perception of increased uterine activity, which can, in some cases, be objectively confirmed by CTG monitoring. It is therefore important to inform pregnant patients undergoing VRL that such subjective and objective uterine responses may occur, but they do not indicate the onset of labor.

Although widely used, the effectiveness of VRL in certain conditions, particularly in adults, remains a topic of ongoing research^6,11,16. While clinical practice often demonstrates positive outcomes, particularly in postural control and pain reduction, further randomized controlled trials are necessary to establish stronger evidence. This consideration is especially relevant when managing chronic musculoskeletal disorders, where the response to VRL may be variable and dependent on patient-specific factors. In our study, during the stimulation of the pressure zones, there were no complications. All patients were able to perform normal activities immediately after stimulation and all CTG records were evaluated as physiological. There is no indication that the use of VRL would threaten the ongoing pregnancy.

Previous research has explored various manual and neuromodulatory approaches to facilitate the onset of labor, including acupuncture, acupressure, reflexology, and other physiotherapeutic methods. Several randomized trials and systematic reviews have reported that acupuncture and acupressure may slightly increase the likelihood of spontaneous labor onset, however, these effects are generally small, inconsistent, and of limited clinical relevance¹⁷. Earlier controlled studies found no significant reduction in induction rates or delivery time compared with standard care¹⁸. A recent meta-analysis of 17 randomized controlled trials involving more than 3,000 women confirmed only a minor increase in spontaneous labor onset compared with no intervention, with no significant difference compared with sham stimulation¹⁹. Similar findings have been reported for reflexology and massage-based interventions, which may influence autonomic activity but lack robust evidence for clinically meaningful effects on labor initiation²⁰.

In this context, Vojta’s reflex locomotion (VRL) can be regarded as part of this broader group of proprioceptive and neuromotor activation techniques. Unlike acupuncture or reflexology, VRL is based on the activation of defined motor reflex patterns through pressure stimulation of specific trigger zones, potentially involving spinal and supraspinal pathways rather than peripheral meridian mechanisms. The present findings therefore extend existing evidence by demonstrating that such neuroreflexive stimulation does not induce preterm or excessive uterine activity, supporting its safety in term pregnancy while highlighting the need for further mechanistic research.

In the official textbook Vojtův princip by Vojta and Peters (3rd edition, 2010)¹, pregnancy is listed as a contraindication for the use of Vojta’s method. The same position is reiterated on the official website of the Vojta Method (www.vojta.com)⁷, which also classifies pregnancy as a contraindication. Our findings, however, do not support pregnancy as a blanket contraindication. In this prospective randomized study, VRL applied in term pregnancies with favourable cervical status was well tolerated, did not induce labor, and was not associated with any observed maternal or fetal complications.

We acknowledge limitations of our study, including a small group of patients and a short period of stimulation. Additionally, we cannot provide an answer regarding the safety of Vojta’s therapy for preterm pregnancies. The study population consisted of women at term with favourable cervical conditions (Bishop score ≥ 6), which was intentionally chosen to minimize obstetric risk in this first prospective evaluation of VRL during pregnancy. This design, however, may have resulted in higher baseline uterine activity and shorter time to spontaneous labour compared with the general pregnant population. Therefore, the results should primarily be interpreted in the context of term pregnancies with cervical readiness, and further research is needed to assess the effects of VRL earlier in gestation or in women with less favourable cervical status.

Furthermore, given the small sample size (n = 40), the study is underpowered to exclude clinically important but rare adverse outcomes, such as tachysystole or transient fetal distress. In addition, neonatal outcomes such as Apgar score, NICU admission, or postpartum complications were not collected, as the study was limited to immediate maternal responses and CTG changes. These neonatal safety parameters should be included as predefined endpoints in future trials. The relatively small sample size did not allow for stratification according to parity or other maternal characteristics. Multiparas and primiparas are known to differ in uterine reactivity at term. Therefore, the absence of stratification may have masked or exaggerated some of the observed effects. As all participants in this pilot study were Caucasian women of Czech origin, the generalizability of the findings to other ethnic or demographic groups may be limited. These factors should be addressed in future studies with larger and more heterogeneous populations. Thus, while our findings are reassuring and indicate no observed harms in this cohort, they should be interpreted as preliminary. Larger randomized trials are warranted to confirm the safety profile of VRL in pregnancy.

The safety of VRL in pregnancies earlier than 37 weeks remains unknown and should be addressed in future research to determine whether gestational age should influence contraindication status.

Conclusion

Our findings indicate that Vojta’s reflex locomotion (VRL) can induce uterine activity in term pregnancy; however, this activity is not sufficient to initiate labor. To our knowledge, this is the first prospective randomized study to objectively evaluate the effect of VRL on uterine activity in pregnant patients. Our findings suggest that the use of VRL in term pregnancy is likely safe and well tolerated. Nevertheless, given the limited evidence, these results should be interpreted with caution and confirmed by further research.

Further research is warranted to evaluate the safety and potential effects of VRL in pregnancies prior to 37 weeks of gestation, in order to establish clear and evidence-based contraindication guidelines for earlier stages of pregnancy.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CTG

Cardiotocograph

DVT

Deep venous thrombosis

PROM

Premature rupture of membranes

VAS

Visual analog scale

VRL

Vojta’s reflex locomotion

TOLAC

Trial of labor after cesarean section

Author contributions

Martinek Milan—Conceptualization, Methodology, Investigation, Figures, Writing—original draft. Zapletal Jan—Conceptualization, Methodology, Investigation, Figures, Writing—original draft. Drochytek Vit—Methodology, Writing—review & editing. Janovska Nikola—Investigation. Svedova Barbora—Investigation. Babkova Anna—Writing - review & editing. Vlk Radovan—Conceptualization, Methodology, Writing—review & editing. Rob Lukas—Writing - review & editing. Novakova Tereza—Conceptualization, Methodology, Investigation, Writing—review & editing. Halaska Jiri Michael—Writing—review & editing.

Funding

This study has been supported by Cooperatio project Maternal and Childhood Care (No. 207035) from the Third Medical Faculty and Physiotherapy and subprogram Physiotherapy and Therapeutic Procedures of the Faculty of the Physical Education and Sport Charles University.

Data availability

The datasets analyzed during this study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics statement and consent to participate

The study was approved by the Ethics Committee of the Faculty Hospital Kralovske Vinohrady (EK-VPl421012021) and retrospectively registered on ClinicalTrials.gov (ID: NCT06339528; registration date: 01/04/2024), URL: https://clinicaltrials.gov/study/NCT06339528. Written informed consent to participate was obtained from all participants prior to inclusion in the study.

Consent for publication

Written informed consent for publication of anonymized clinical data was obtained from all participants.