Ice Massage for the Reduction of Labor Pain

Bette L. Waters, CNM, RN, Jeanne Raisler, CNM, DrPH, FACNM


J Midwifery Womens Health. 2003;48(5) 

In This Article


A one-group, pretest, posttest design was chosen.[20] The pretest was a 100-mm VAS. The VAS has been extensively used and validated in pain research and is considered to be a valid measure, especially in a one-time intervention study.[21,22] The pretest was used to measure labor pain intensity before ice massage and served as the control. Posttest 1 was a 100-mm VAS for both the right hand and the left hand. It was used to measure pain intensity during ice massage intervention and to compare pain intensity on the right hand versus the left hand. Pretest and posttest 1 scores were compared by using standard analysis of variance, Statistical Analysis System, Version 8.00. Pain response differences were multiples that consisted of three elements: pain before massage, pain during massage of the left hand, and pain during massage of the right hand. These differences were identified by using Duncan's New Multiple Range test.[23]

Posttest 2 was the McGill Pain Questionnaire (MPQ) Verbal Rating Scale. The MPQ is the most widely used instrument in pain research and practice.[24,25] It consists of verbal pain descriptors designed to capture the intensity continuum of pain ranging from mild to excruciating, and ranked numerically from 1 to 5. The participant scored two questions using the MPQ: (1) What was your pain before you started the ice massage? (2) What was your pain while using the ice massage? The MPQ was analyzed by using a standard analysis of variance equivalent to a paired t test.[26] If data collected in posttest 2 follows the pattern of posttest 1, it is regarded as a valuable corroboration of the data.

To reduce threats to the validity, posttest 1 was administered immediately after the intervention, 40 minutes or less after scoring the pretest. This small window of time between the pre-and posttest helped to eliminate intervening events that could alter the posttest scoring. Posttest 1 was presented to the study participants on a separate sheet of paper so they could not see where they had marked on the pretest VAS. VAS tool copies were made from a master copy on the same copy machine. Posttest 2 was administered within 24 hours after the delivery. Postponing the scoring of this test until after the woman, especially the primigravida, had experienced the process of the labor and birth with full knowledge of all its intensity strengthens any corroborative data.

Study participants were a convenience sample of English-speaking Hispanic and white Medicaid recipients who received prenatal and delivery care from a clinic team of certified nurse-midwives and obstetricians at a 250-bed hospital in New Mexico. They were recruited for the study on the basis of the availability of the investigator after being evaluated and admitted to the hospital in labor by their midwife or physician. The gestational age of the participants was determined to be between 37 and 41 weeks by an early sonogram. All participants had a reactive fetal heart rate monitoring strip and were having contractions at least every 10 minutes with some cervical changes, either effacement or dilation. Women diagnosed with pre-eclampsia or chorioamnionitis were excluded from the study. Women whose labor was induced, those who had narcotics in the past 8 hours, and women with an underlying disease that precluded attendance by a nurse-midwife were excluded from the study. In addition, women dilated more than 8 cm were excluded. It was believed that the intensity of the labor contractions during this transition could decrease participants cognitive abilities and thus compromise the data obtained. Previous research reported[27] that women found the VAS difficult to use when experiencing severe labor pain. The use of women in the early stages of labor eliminated the ethical issue of withholding pain medications they might want to use as labor progressed.

The sample size was not predetermined. The goal was to recruit as many subjects as possible within a 12-month period. Unlike many other studies[28] of non-pharmacological methods of pain relief, these participants had no prior commitment to giving birth without pain medication, and most expected to receive either narcotic or epidural analgesia at some point in the labor.

The clinical investigation protocol and consent form were approved and monitored by Memorial Medical Center's Institutional Review Board. After obtaining verbal and written informed consent, the investigator presented the pretest and explained to the subject how to mark her pain intensity at the present moment on the VAS. Ice massage was started at the initiation of the next contraction. Approximately one-third cup of crushed ice was placed in the center of a soft, thin terry wash cloth, and the four corners of the washcloth were lifted to the center and twisted to make a small ice bag. The ice bag fit snugly between the thumb and forefinger. To ensure that cold was applied only to the skin of the palm, the ice bag was placed on the medial (palm side) aspect of the hand (Figure 3).

Ice bag use at large intestine energy point 4 (LI4). Correct positioning of small ice bag for massage stimulation of LI4.

The lateral aspect of the participant's hand was supported by the hand of the person performing the massage. The massage was stopped when the contraction ended and restarted when the next contraction began. The ice bag was rocked back and forth over the area of the web of skin between the thumb and the forefinger. The pressure of the ice bag was comparable to light scratching and was intended to mildly irritate the neuron endings in the skin

It should be noted that the exact point of LI4 is located on the medial aspect of the first metacarpal. The skin or epidermis located directly over this point is part of the outer part of the hand and is thin. Ice massage over this area can cause breakdown of skin integrity due to cold temperatures and friction. However, the web of skin between the thumb and forefinger shown in Figure 2 is part of the thick, hard, and horny texture of the palm and can withstand the intermittent friction and cold temperatures used in this technique.[29]

The massage was carried out on one hand for 20 minutes or throughout three or four contractions, whichever occurred first. It was then repeated in the same manner on the other hand. The selection of right hand or left hand first was based on what activity the study participant was engaged in at the time the ice massage began. If she was in bed, the choice was determined by which side of the bed the fetal monitor was located. If she was walking in the halls or soaking in the Jacuzzi, the investigator accepted the participants choice.

At the end of the massage period (40 minutes or less), the subject was given VAS posttest 1 on a separate sheet of paper. She marked on the VAS the amount of pain she experienced while using the ice massage on the right hand and the left hand. At the end of the intervention, the investigator taught a family member how to continue the ice massage if the study participant desired.

Posttest 2, the ranked MPQ designed to measure memory of pain, was completed by the participant within 24 hours after delivery.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.