NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Likis FE, Andrews JC, Collins MR, et al. Nitrous Oxide for the Management of Labor Pain [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2012 Aug. (Comparative Effectiveness Reviews, No. 67.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Nitrous Oxide for the Management of Labor Pain

Nitrous Oxide for the Management of Labor Pain [Internet].

Show details


State of the Literature

We identified a total of 59 distinct studies reported in 58 publications: two of good quality; 11 fair; and 46 poor. Thirty-three percent of the studies identified were randomized clinical trials (RCTs), a smaller proportion were clinical trials without clear evidence of randomization (7%), and the balance are observational research.

Strength of Evidence

Overall the strength of evidence to answer the Key Questions (KQ) was insufficient for effectiveness in managing labor pain (KQ1), effect on route of birth (KQ3), and health system factors (KQ5); low for satisfaction (KQ2); and moderate for harms (KQ4) (Table 12). Deficiencies in the strength of evidence most often related to a preponderance of study designs with high risk of bias; inconsistent findings across studies and inconsistencies among outcomes that would be expected to show corresponding benefit; use of intermediate outcomes; and small studies with poor precision. In the summary below, we provide strength of evidence ratings by Key Question.

Table 12. Strength of evidence for nitrous oxide for the management of labor pain.

Table 12

Strength of evidence for nitrous oxide for the management of labor pain.

Principal Findings and Considerations

KQ1. Effectiveness of Nitrous Oxide for Labor Pain Management

Twenty-one studies addressed the effectiveness of nitrous oxide using some measurement of pain or pain relief.38, 40-41, 44-45, 47-49, 51, 53-54, 58, 62, 66, 68-70, 75-76, 78-79 Four studies were of fair quality,69-70, 75-76 and 17 were of poor quality.38, 40-41, 44-45, 47-49, 51, 53-54, 58, 62, 66, 68, 78-79 There was considerable variation across studies in many aspects including the concentration of nitrous oxide and frequency (continuous vs. intermittent) administered, additional pain management methods used, and methods and persons (i.e., women, obstetricians, midwives, and anesthesia providers) assessing pain and pain relief. The substantial variation in timing of assessment may have affected the reported outcomes because women's opinions about pain relief change with time lapsed after birth.53-54, 66

The majority of the effectiveness studies (12 of 21) had as comparators other inhalational anesthetic gases that are not used to manage labor pain in the United States. Only one study compared nitrous oxide with placebo and found no significant difference in pain scores. As expected, epidurals provide more effective pain relief than nitrous oxide. It may be counterproductive to evaluate pain scores, which require focusing on the level of pain, in women using nitrous oxide, which is intended to produce dissociation from pain. What these studies are unable to demonstrate is whether nitrous provided adequate pain relief for women who knowingly accept less effective pain relief in exchange for increased mobility, less intervention and monitoring, and avoidance of potential complications associated with epidurals. Generally speaking, therefore, pain relief is likely to be an inadequate measure of effectiveness for nitrous oxide in the absence of other outcomes such as women's satisfaction.

KQ2. Effect of Nitrous Oxide on Women's Satisfaction

Nine studies addressed women's satisfaction with their birth experience or pain management.40, 45, 49, 56, 58, 62, 71, 74, 79 One study was of good quality,56 one of fair quality,74 and seven of poor quality.40, 45, 49, 58, 62, 71, 79 Measurements of satisfaction were not uniform making it impossible to synthesize studies. Satisfaction may be a more relevant measure of effectiveness than assessment of pain because nitrous oxide is not intended to provide complete pain relief.

KQ3. Effect of Nitrous Oxide on the Route of Birth

Six studies compare the route of birth in women who used nitrous oxide with women who used other pain management methods.40, 54, 56, 58, 62, 66 Two of these only included women who had vaginal births,40, 54 and five were of poor quality.40, 54, 58, 62, 66 The strength of the evidence is insufficient to determine the effect of nitrous oxide on the route of birth.

KQ4. Adverse Effects of Nitrous Oxide for Labor Pain Management

Forty-nine studies addressed the maternal, fetal, neonatal, and occupational harms related to nitrous oxide use during labor.12-13, 16, 26-30, 32-43, 45-61, 63-69, 72-73, 77, 80 Two were of good quality,36, 56 seven of fair quality,26, 30-32, 42, 69, 72 and forty of poor quality.12-13, 16, 27-29, 33-35, 37-41, 43, 45-55, 57-61, 63-68, 73, 77, 80 Although these 49 studies report data from more than 27,000 women, only six of these studies were conducted in the United States (n = 2,445 women). In addition, one-third (16 of 49) of studies reporting harms were conducted prior to 1980 when nitrous oxide was often used in combination with sedatives, tranquilizers, and other inhaled anesthetics in labor, a practice that has largely been abandoned. Studies reporting harms associated with sedative analgesic regimens may not translate effectively to contemporary labor analgesia practice. For example, in older studies amnesia in labor was considered to be a positive outcome.

Most maternal harms reported in the literature were unpleasant side effects that affect tolerability (e.g., nausea, vomiting, dizziness, and drowsiness). Some maternal harms (e.g., nausea and oxygen desaturation) are common in all laboring women regardless of the type of analgesia used. Study sizes were inadequate to assess for unusual or rare harms that might be more serious in terms of morbidity.

Nitrous oxide is transmitted via the placenta and is rapidly eliminated by the neonate following birth once breathing begins. Apgar scores in newborns whose mothers used nitrous oxide did not differ significantly from those of newborns whose mothers used other labor pain management methods or no analgesia. Followup of newborns was short, most frequently lasting only to birth or discharge of the neonate from the hospital.

Limited data on occupational harms are available thus it is difficult to draw conclusions regarding potential occupational harms as a result of exposure to nitrous oxide. Evidence about occupational levels of nitrous oxide is limited, and some studies were conducted prior to the use of room ventilation systems or scavenging systems. The implementation of these systems in clinical practice has reduced occupational exposure, which should mitigate potential risks of exposure.

KQ5. Effects of Provider and Health System Factors

No studies addressed KQ5. It is discussed as a part of future research.


Applicability describes the extent to which study populations and characteristics in the literature reviewed apply to the larger population. In this report, the study populations were healthy women in labor who should be similar to the target population. The eligibility criteria and participant characteristics were not always explicitly detailed. Some participants were excluded due to choice of alternate pain management methods.

Most studies used a 50/50 mix of nitrous and oxygen, often premixed in the form of Entonox®. The 50/50 mix is available, although Entonox is not used in the United States and has not been reviewed by the U.S. Food and Drug Administration. In addition, mechanical equipment for administration of nitrous oxide in labor and delivery has very limited availability in the United States at the time of this writing. In the studies related to harms (Key Question 4), the intervention varied significantly in terms of dose, frequency, and duration. In many studies participants received unspecified amounts of narcotics and/or sedating agents. Studies prior to 1980 are not applicable to current guidelines for clinical use.

The comparators include standard pain management methods, such as epidural, narcotics, and nonpharmacologic methods such as transcutaneous electrical nerve stimulation (TENS). However, some comparators are not commonly used and/or available for laboring women, such as other inhalational anesthetic gases.

For KQ1, the most frequent outcome was an assessment of pain, generally during labor. Some studies retrospectively assessed pain in the immediate postpartum period and/or weeks to months after birth. The methods of pain assessment were heterogeneous. Those assessing outcomes included participants, obstetricians, midwives, and anesthesia providers. Satisfaction with pain management and the birth experience, as reported by the women were the outcome measures for KQ2. The outcomes for KQ3 were vaginal birth, assisted vaginal birth, and cesarean. None of the studies had a cesarean birth rate greater than 10 percent, which is much lower than the most recently reported U.S. rate of 32 percent.1 For KQ4, the most frequent outcomes were assessments of nausea, vomiting, dizziness, drowsiness, hypoxia, oxygen saturation, Apgar scores, and cord blood gases.

Only six of 58 studies were conducted in the United States. The options for labor pain management in the United States are somewhat dissimilar to those in other countries because nitrous oxide for laboring women is widely available outside of the United States, whereas in this country its availability is extremely limited. While setting was not a criterion for inclusion or exclusion, all of the studies were conducted in hospitals. Thus the effectiveness, women's satisfaction, route of birth, and harms associated with nitrous oxide in birth centers and the home setting have not been reported.

Future Research

State of the Science

Nitrous oxide has been used for labor pain management since the 1930s, primarily outside the United States.4 Much of the literature on this topic is older with nearly half of the studies in this review published prior to 1990 and one-quarter before 1980. Over the past decade, there has been growing interest in the use of nitrous oxide for laboring women in the United States. As use of nitrous oxide for labor pain management increases, continued research is warranted. Topics that would benefit from consideration include:

Methodologic Priorities

  • Clearly documenting the mix of nitrous oxide used and the timing and mode of administration.
  • Performing studies that use doses and equipment consistent with contemporary U.S. maternity care.
  • Developing outcome measures that assess effectiveness as defined by women choosing nitrous oxide.
  • Using standardized and validated outcome measures to assess pain and women's satisfaction.
  • Including women's assessment of pain, rather than only providers', in all studies that report this outcome.
  • Performing qualitative research in addition to quantitative studies.
  • Conducting sequential analysis trials in which women can opt-in and opt-out of nitrous oxide.
  • Conducting studies in out-of-hospital birth settings (i.e., freestanding birth centers and home births).
  • Building consensus about critical maternal, fetal, neonatal, childhood, and occupational exposure outcomes, developing a minimal core data set for future research.
  • Designing human studies that examine apoptosis, which has been observed in rodents exposed to high doses of systemic anesthetics.
  • Developing electronic medical record approaches to long-term surveillance for adverse effects.

Content Priorities

  • Exploring anti-anxiety effects of nitrous oxide during labor.
  • Examining the influence of nitrous oxide on whether and when women choose to use other labor pain management methods.
  • Investigating the impact of nitrous oxide on use of cointerventions, route of birth, maternal-newborn bonding, and breastfeeding.
  • Assessing fetal/neonatal clearance of nitrous oxide.
  • Determining optimal methods for minimizing occupational exposures, such as room ventilation and scavenging measures.
  • Assessing potential occupational harms, including nitrous oxide abuse and addiction.
  • Identifying health system factors influencing the use of nitrous oxide for the management of labor pain, including but not limited to provider preferences, availability, setting, and resource utilization.
  • Determining provider and patient education needed for nitrous oxide use in labor,
  • Analyzing cost effectiveness of nitrous oxide and other labor pain management methods.

Current and Future Research

Recently completed and ongoing research includes the following:


  • One study on the effect of labor analgesia on babies' movement after birth, with nitrous oxide use or no analgesia as the control group.


  • Zero studies.


  • One study on the comparison of the effects of Entonox and TENS in labor pain.


The literature addressing nitrous oxide for the management of labor pain has few studies of good or fair quality. Synthesis of effectiveness and satisfaction studies was challenging because of heterogeneous interventions, comparators, and outcome measures. Satisfaction may be a more relevant measure of effectiveness than assessment of pain because nitrous oxide is not intended to provide complete pain relief. The strength of evidence for the effect of nitrous oxide on route of birth was insufficient. Most maternal harms reported in the literature were unpleasant side effects that affect tolerability (e.g., nausea, vomiting, dizziness, and drowsiness), and Apgar scores did not differ significantly across labor pain management methods. Data for occupational harms were limited. Research assessing nitrous oxide is needed across all of the Key Questions examined: effectiveness, women's satisfaction, route of birth, harms, and health system factors affecting use.


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.8M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...