Presented at the Society of Academic Emergency Medicine annual meeting, New Orleans, LA, May 2009.
Heat or Cold Packs for Neck and Back Strain: A Randomized Controlled Trial of Efficacy
Version of Record online: 23 APR 2010
© 2010 by the Society for Academic Emergency Medicine
Academic Emergency Medicine
Volume 17, Issue 5, pages 484–489, May 2010
How to Cite
Garra, G., Singer, A. J., Leno, R., Taira, B. R., Gupta, N., Mathaikutty, B. and Thode, H. J. (2010), Heat or Cold Packs for Neck and Back Strain: A Randomized Controlled Trial of Efficacy. Academic Emergency Medicine, 17: 484–489. doi: 10.1111/j.1553-2712.2010.00735.x
- Issue online: 23 APR 2010
- Version of Record online: 23 APR 2010
- Received August 20, 2009; revision received October 25, 2009; accepted November 15, 2009.
- back injuries;
- neck injuries;
- pain measurement;
Objectives: Acute back and neck strains are very common. In addition to administering analgesics, these strains are often treated with either heat or cold packs. The objective of this study was to compare the analgesic efficacy of heat and cold in relieving pain from back and neck strains. The authors hypothesized that pain relief would not differ between hot and cold packs.
Methods: This was a randomized, controlled trial conducted at a university-based emergency department (ED) with an annual census of 90,000 visits. ED patients >18 years old with acute back or neck strains were eligible for inclusion. All patients received 400 mg of ibuprofen orally and then were randomized to 30 minutes of heating pad or cold pack applied to the strained area. Outcomes of interest were pain severity before and after pack application on a validated 100-mm visual analog scale (VAS) from 0 (no pain) to 100 (worst pain), percentage of patients requiring rescue analgesia, subjective report of pain relief on a verbal rating scale (VRS), and future desire for similar packs. Outcomes were compared with t-tests and chi-square tests. A sample of 60 patients had 80% power to detect a 15-mm difference in pain scores.
Results: Sixty patients were randomized to heat (n = 31) or cold (n = 29) therapy. Mean (±standard deviation [SD]) age was 37.8 (±14.7) years, 51.6% were female, and 66.7% were white. Groups were similar in baseline patient and pain characteristics. There were no differences between the heat and cold groups in the severity of pain before (75 mm [95% CI = 66 to 83] vs. 72 mm [95% CI = 65 to 78]; p = 0.56) or after (66 mm [95% CI = 57 to 75] vs. 64 mm [95% CI = 56 to 73]; p = 0.75) therapy. Pain was rated better or much better in 16/31 (51.6%) and 18/29 (62.1%) patients in the heat and cold groups, respectively (p = 0.27). There were no between-group differences in the desire for and administration of additional analgesia. Twenty-five of 31 (80.6%) patients in the heat group and 22 of 29 (75.9%) patients in the cold group would use the same therapy if injured in the future (p = 0.65).
Conclusions: The addition of a 30-minute topical application of a heating pad or cold pack to ibuprofen therapy for the treatment of acute neck or back strain results in a mild yet similar improvement in the pain severity. However, it is possible that pain relief is mainly the result of ibuprofen therapy. Choice of heat or cold therapy should be based on patient and practitioner preferences and availability.
ACADEMIC EMERGENCY MEDICINE 2010; 17:484–489 © 2010 by the Society for Academic Emergency Medicine
Acute neck and back strains are very common presentations in the emergency department (ED). Topical application of heat or cold packs is a common adjunct in the treatment of such injuries. Application of cold packs or heating pads is a simple, inexpensive, and widely available approach to alleviating the discomfort associated with acute sprains and strains. Numerous reports advocate the use of heat and/or ice for the treatment of soft tissue injury and muscle spasms, as well as a variety of other painful conditions.1–4 However, there is very little evidence supporting these recommendations.
A number of studies have evaluated the analgesic effects of heat and cold therapy separately; however, few if any studies have directly compared their efficacy in acute strains. Our goal was to compare the efficacy of hot and cold packs in relieving pain from acute neck and back strains in the ED. We hypothesized that both therapies would result in a significant reduction in pain, defined as a 15-mm difference in visual analog scale (VAS) scores at 30 minutes.
This was a single-center, prospective, randomized clinical trial comparing the effect of heat versus ice in the treatment of acute neck or back strains. The study was approved by our institutional review board and registered with ClinicalTrials.gov prior to inception. Informed written consent was obtained from all patients.
Study Setting and Population
The study was conducted in a university-based ED with 90,000 visits per year from April to July 2008. A convenience sample of ED patients were identified by trained research assistants5 who received instruction on the study criteria and objectives. Research assistants were present in the ED for patient enrollment Monday through Saturday from 8 am until 10 pm. Research assistants identified patients greater than 18 years of age presenting to the ED with a complaint of neck or back pain. Patients were eligible for the study if neck or back pain was less than 24 hours’ duration and resulting from a minor injury (lifting, pulling, twisting, or minor motor vehicle collision). Patients were excluded from study enrollment if the pain was an exacerbation of a prior neck or back injury or if they had any signs, symptoms, or history of radiculopathy, neurologic deficits, or spinal fracture. Patients were also excluded if they had a contraindication to treatment with nonsteroidal anti-inflammatory medications such as prior allergy, medication intolerance, or history of peptic ulcer disease.
Collection of demographic and clinical data was performed by a trained research assistant using a standardized data collection form. Data collected included demographic information such as age, sex, and race. Specific historical information included the quality and location of pain, mechanism and duration of injury, and medication used prior to presentation. Patients were asked to report the level of discomfort by placing a vertical mark on a 100 mm horizontal VAS marked no pain and worst pain at the low and high ends of the scale.6 Following initial data collection, all patients were given 400 mg of ibuprofen orally.
Research personnel not involved in any other aspect of the study generated treatment assignments with a computerized random-number program using random allocation procedure. Treatment assignments were enclosed in sequentially numbered, opaque, sealed envelopes. The envelopes contained even proportions of the two study treatments. For each enrolled patient, the next in the series of envelopes was opened to reveal the treatment allocation. Patients were randomized to a 30-minute continuous treatment with either Instant Cold Pack (Dynarex Corp., Orangeburg, NY) or an electric heating pad. Prior to study initiation, we determined skin temperature for each treatment modality. The electric heating pad set on high maintained an average skin temperature of 132°F, varying between 130 and 135.6°F. The instant cold pack maintained an average skin temperature of 28.7° F, varying between 19.9° and 34.1°F. Due to the nature of the intervention, it was not possible to blind the patient and physician to the treatment allocation.
Immediately after treatment, patients completed a second data collection form, blinded to the responses on the initial data collection form. Patients were asked to rate the level of pain by placing a mark on a 100-mm VAS and rate the change in pain on a verbal rating scale (VRS): much better, a little better, unchanged, a little worse, or much worse. Data on the desire for additional analgesics and patient satisfaction with thermal therapy were also collected.
Outcomes and Measurements
The main outcome measure was the difference in pain severity on the 100-mm unhatched VAS 30 minutes after application of the cold or heat pack. This was calculated by subtracting the pain score after therapy from the pain score prior to therapy. Secondary outcome measures included the percentage of patients requiring rescue analgesia, degree of pain relief, and future desire for similar packs.
All data were entered into SPSS 16.0 for Windows (SPSS Inc., Chicago, IL). Categorical variables are presented as the percentage frequency of occurrence, and study groups were compared using chi-square tests or the Fisher’s exact test, as appropriate. Continuous variables are presented as means with standard deviations (SDs) or 95% confidence intervals (CIs), and comparisons were performed using t-tests after ascertaining the normality of the data. Categorical variables are presented as percentages, and comparisons were performed using chi-square tests or Kruskal-Wallis test. A sample of 60 patients (30 patients in each group) provided an 80% power to detect a difference of 15 mm on the VAS. A value of α = 0.05 was used to determine the level of significance with no adjusting for multiple comparisons.
A convenience sample of 60 patients was randomized to heat (n = 31) or cold (n = 29) therapy. The mean (±SD) age of the patients in the study was 37 (±13) years. Sixty-four percent were Caucasian, 12% black or African American, 10% Hispanic, and 8.5% Asian. The majority of patients reported lower back pain (61%), followed by neck pain (27%) and upper back pain (12%). The pain was sudden in onset in 62% of cases and constant in intensity in 80% of the patients. Overall, 17 patients (28%) self-treated with medication prior to the ED presentation: five (8%) took acetaminophen, five (8%) took a nonsteroidal anti-inflammatory drug (NSAID), and seven (12%) took a narcotic analgesic. The mean (±SD) time of medication pretreatment was 15 (±6) hours. Groups were similar in baseline demographic and clinical characteristics (Table 1).
|Characteristics||Heat (n = 31)||Cold (n = 29)||p-value|
|Age, yr (SD)||38 (±15)||36 (±11)||0.66|
|Sex, n (% female)||16 (52)||14 (48)||0.8|
|Location of pain, n (%)|
|Neck||4 (13)||11 (38)|
|Upper back||6 (19)||1 (3)||0.62|
|Lower back||19 (61)||15 (52)|
|Missing data||2 (7)||2 (7)|
|Mean pretreatment VAS, mm (95% CI)||75 (66–83)||72 (65–78)||0.56|
|Mean post treatment VAS, mm (95% CI)||66 (57–75)||64 (56–73)||0.75|
There were no differences in VAS pain scores between the heat and cold groups both before and after treatment (Table 2). The mean decrease in pain scores was also similar in the heat and cold groups (9 [±16] mm vs. 8 [±10] mm, respectively). Although statistically significant, the decrease in each group did not achieve the predefined clinically significant reduction in pain. The mean VAS difference in patients reporting pain as “a little better” or “much better” was 13 mm. The patient’s subjective impression of pain relief as reported on a VRS is illustrated in Figure 1. There was no statistical difference in VRS reporting between groups (p = 0.27).
|Outcomes||Heat (n = 31)||Cold (n = 29)||Difference (95% CI)||p-value|
|VAS difference, mm (SD)||9 (±16)||8 (±10)||1 (−5.7 to 7.9)||0.75|
|Requested rescue medication, n (%)||18 (58)||12 (41)||17 (−8 to 39)||0.2|
|Administered rescue medication, n (%)||10 (32)||7 (24)||8 (−14 to 29)||0.49|
|Patient satisfaction, n (%)||25 (81)||22 (76)||5 (−16 to 25)||0.65|
Additional analgesia was desired by 18/31 (58.1%) and 12/29 (41.4%) patients in the heat and cold therapy groups, respectively (p = 0.24). Adjunctive analgesia was administered to 10/31 (32.3%) and 7/29 (24.1%) patients treated with heat or cold, respectively (p = 0.48). When asked, 25/31 (80.6%) patients in the heat group and 22/29 (75.9%) patients in the cold group would use the same therapy if injured in the future (p = 0.65).
Clinicians treating musculoskeletal injuries are frequently asked about the benefits and timing of ice versus heat therapy. The results of our study demonstrate that a 30-minute application of heat or cold does not significantly improve the pain of acute neck or back strain. However, of the patients reporting improvement of symptoms on a verbal scale, the mean change in VAS exceeded the minimal clinically significant difference for pain proposed by Todd et al.7
Heat and cold packs are inexpensive, readily available adjuncts in the treatment of musculoskeletal disorders. Application of heat or cold may alleviate the pain of muscle strain through a number of physiologic mechanisms. The pain threshold may be increased by application of heat or cold.8,9 Cold has been demonstrated to decrease nerve conduction velocity, exerting a local anesthetic effect.10 Topical heat or cold acts as a counterirritant,11 increasing small nonmyelinated C-fiber activity that inhibits nonciceptive signals within the spinal cord and brain stem.12 Lowering of muscle temperature reduces muscle spindle sensitivity and overall muscle tone.13 Heating decreases gamma-fiber activity in muscle, reducing the sensitivity of muscle spindle to stretch.14 Heat may also stimulate areas of the brain, exerting a psychosomatic effect.
A joint clinical practice guideline from the American College of Physicians (ACP) and the American Pain Society recommends application of heat by heating pads or heated blankets as a self-care option for short-term relief of acute low back pain.15 A survey of physical therapists found that 81.9 and 65.7% believe that cold and heat are effective in the management of most patients with acute low back pain.16 The number of published studies evaluating the efficacy of heat or cold treatment for acute low back pain is limited. A recent Cochrane review reports moderate evidence supporting the use of heat wrap therapy and insufficient evidence for cold treatment of acute low back pain.17 The Cochrane review was unable to determine differences between heat and cold therapy for low back pain because of the conflicting evidence and the heterogeneous nature of existing trials.
There are no systematic comparisons of heat versus cold as an adjuvant therapy in the ED treatment of acute neck or back strain. Landen18 evaluated the effect of heat versus ice for the treatment of a mixed population of patients with low back pain (acute, subacute, and chronic). In 53 patients hospitalized for acute pain (<48 hours onset), there was no difference in pain response between hot packs and ice massage. However, patients allocated to hot pack treatment had a shorter length of stay (4.08 days vs. 5.55 days). We found a similar pain outcome; neither heat nor cold application resulted in a significant change in pain severity.
It is interesting to note that the majority of patients would utilize a form of thermal therapy for future injuries. In light of our findings, this supports the expectancy theory of placebo effect, the patient’s belief or expectation that the placebo intervention is effective.19 It is also interesting that our findings further support the concept of oligoanalgesia in the ED.20 Additional analgesia was desired by 30 patients yet only provided to 17, slightly more than half. This suggests the need for continued education on oligoanalgesia.
The results of our study favor the null hypothesis: no difference exists between therapies. As such, there are many potential biases that may have contributed to a Type 2 error. The lack of blinding may have created an expectation bias. Due to the nature of the study, it was difficult to blind patients to treatment allocation. Knowledge of the exposure may have influenced outcomes adversely. Although larger treatment effects can result when appropriate levels of blinding are not utilized,21 our study demonstrates the opposite. We believe that expectation bias did not substantially contribute to the study outcome because most of the patients reported satisfaction with treatment allocation.
The study did not include a control group. In theory, control groups should eliminate alternate explanations of experimental results. The ideal scientific study would include both a positive control group and a negative control group. The positive control group minimizes false-negative results, confirming that the variable (in our case, heat or cold) is effective. The positive control group in our study would have been a group that received ibuprofen only. A negative control group should provide a negative result, minimizing false positives. The negative control group in our study would have been a group receiving placebo ibuprofen and no thermal therapy. There is compelling evidence that NSAIDs or acetaminophen is effective for the short-term relief of patients with low back pain.22 The ACP guideline recommends acetaminophen or NSAIDs as a first-line medication option for patients with low back pain.15 Given the data, we did not feel that it was ethical to have a negative control group. However, it would have been reasonable to include a positive control group receiving ibuprofen only. As a result, the design does not provide the ability to differentiate the effects of heat vs. cold alone and does not prove the additional benefit of these modalities beyond ibuprofen. It is possible that the magnitude of our findings is limited by not measuring the natural course of pain in such injuries.
Enrollment criteria were based on clinical suspicion for acute neck or back strain. We did not apply specific clinical criteria to differentiate muscle strain from other etiologies of neck or back pain. Furthermore, the experimental design consisted of a convenience sample of patients. Convenience sampling is a type of nonprobability sampling that lends bias to the study because participants select themselves. Because this form of sampling is less representative of the total population, it is difficult to generalize the results. However, we have no reason to believe that this population would respond differently from a random sample.
The therapy was prescribed for 30 minutes’ duration based on previous literature pertaining to muscle cooling and warming. However, treatments of short duration may not have sufficient time for a noticeable effect in the outcome of interest (timing bias).23 Several studies have demonstrated that continuous heat therapy was effective in alleviating the pain from acute low back pain.24–26 Nadler et al.24 concluded that 8 hours of continuous heat therapy (104°F) provided higher pain relief scores compared to oral acetaminophen or oral ibuprofen as a first-line therapy for acute muscular low back pain. Mayer et al.25 demonstrated that a combination of 8 hours’ continuous low-level heat therapy (104°F) with or without exercises provided higher pain relief scores than exercise alone or an educational booklet on acute low back problems. Our findings would not extrapolate to the repeated or continuous use of thermal therapy for acute neck or low back stains.
The literature suggests that muscle relaxation and pain severity may be directly related to the change in muscle temperature.27 In a study on patients with chronic back pain, Roberts et al.28 speculated that cold and heat packs were not as effective as ice massage in relieving pain because the slow cooling and heating process allows for adaptation and reduces peripheral spinal cord input. Additionally, body mass index may be as important a variable as duration of treatment. Johnson et al.29 found a significant correlation (r = 0.81) between body mass index and intramuscular temperature in patients undergoing cold submersion. We did not measure body mass index, nor did we alter the treatment duration based on body habitus.
It is possible that heat or cold may not have a direct impact on pain severity, but may assist in time to recovery. Nadler et al.26 demonstrated that application of a continuous low-level heat wrap for 48 hours has long-lasting effects on low back pain, flexibility, and stiffness. Nuhr et al.30 speculated that early external warming of patients with acute low back pain (<6 hours) may have beneficial long-term effects. We did not collect follow-up data to determine ongoing pain severity, use of analgesic medication, or time to recovery.
The addition of a 30-minute topical application of a heating pad or cold pack to ibuprofen therapy for the treatment of acute neck or back strain results in a mild yet similar improvement in the pain severity. However, it is possible that pain relief is mainly the result of ibuprofen therapy. Choice of heat or cold therapy should be based on patient and practitioner preferences and availability.
- 8Ch 12 Application of heat and cold in the clinical setting. In: LehmannJF. Therapeutic Heat and Cold, ed 4. Baltimore, MD: Williams and Wilkins, 1990, pp 633–5., .
- 13Cryotherapy. In: LehmannF (ed.). Therapeutic Heat and Cold, ed. 4. Baltimore, MD: Williams and Wilkins, 1990, pp 590–632., .
- 14Physiologic responses to heat and cold. In: LichtS (ed). Therapeutic Heat and Cold, ed. 2. Baltimore, MD: Waverly Press, 1965, pp 126–169., .
- 17Superficial heat or cold for low back pain. Cochrane Database Syst Rev 2006; 1:CD004750., , , , .
- 27Physiatric approach to chronic pain. in: AronoffGM (ed). Evaluation and Treatment of Chronic Pain. 2nd ed. Baltimore, MD: Williams and Wilkins, 1992, pp 176–183..
- 28Relief of chronic low back pain: heat versus cold. In: AronoffGM (ed). Evaluation and treatment of chronic pain. 2nd edition. Baltimore, MD: Williams and Wilkins, 1992, pp 299–301., , , et al.