Quantitative Critique

Copyright © 2017 by National Association of Orthopaedic Nurses. Unauthorized reproduction of this article is prohibited.

344 Orthopaedic Nursing • September/October 2017 • Volume 36 • Number 5 © 2017 by National Association of Orthopaedic Nurses

C old therapy is the therapeutic application of a substance to reduce tissue temperature, blood flow, and cellular metabolism. Local applica- tion of cold produces vasoconstriction, which

reduces cutaneous blood flow, thus decreasing the local inflammatory response and edema formation (Bleakley, McDonough, & MacAuley, 2004; Levy & Marmar, 1993). The anesthetic effect of local cooling is produced by slowing or eliminating the transmission of pain signals. Low temperatures impede neuromuscular transmis- sions and thus raise pain thresholds (Abramson, Chu, & Tuck, 1966). Cooling of the muscle tissue inhibits activity of the muscle spindles and reduces spasm (Airaksinen

BACKGROUND: Cold therapy used in the sports medicine settings has been found to be effective in reducing postop- erative pain; however, there are limited studies that examine the effect of cold therapy on postoperative pain in patients with posterior lumbar spinal fusion. PURPOSE: The purpose of this study was to determine the effects of cold on postoperative spine pain and add to the body of knowledge specific to practical application of cold therapy in the spine surgery setting. METHODS: Researchers used a two-group randomized control design to evaluate the effects of local cold therapy on postoperative pain and analgesia use after lumbar spinal fusion surgery. The primary outcome was postoperative pain. Secondary outcomes included analgesia use and per- ceived benefit of cold therapy. RESULTS: The intervention (cold) group had a marginally greater reduction in mean Numerical Rating Scale score across all 12 pain checks (M ± SD = −1.1 ± 0.8 points reduction vs. −1.0 ± 0.8 points reduction, p = .589). On average, the intervention group used less morphine equiva- lents (M ± SD = 12.6 ± 31.5 vs. 23.7 ± 40.0) than the control group across pain checks seven to 12 (p = .042). CONCLUSIONS: This study provides additional evidence to support the use of cold therapy as an adjuvant pain man- agement strategy to optimize pain control and reduce opi- oid consumption following spine fusion surgical procedures.

Effects of Localized Cold Therapy on Pain in Postoperative Spinal Fusion Patients A Randomized Control Trial

Patricia Quinlan ▼ Jack Davis ▼ Kara Fields ▼ Pia Madamba ▼ Lisa Colman ▼ Daniela Tinca ▼ Regina Cannon Drake

Patricia Quinlan, PhD, MPA, RN, CPHQ, AVP Nursing Excellence, Department of Nursing, Hospital for Special Surgery, New York, NY.

Jack Davis, MSN, RN, ONC, Manager, Patient Education Programs and Research, Department of Nursing, Hospital for Special Surgery, New York, NY.

Kara Fields, MS, Statistical Analyst, Healthcare Research Institute, Hospital for Special Surgery, New York, NY.

Pia Madamba, BSN, RN, ONC, Patient Care Director, Department of Nursing, Hospital for Special Surgery, New York, NY.

Lisa Colman, RN, ONC, Clinical Nurse, Department of Nursing, Hospital for Special Surgery, New York, NY.

Daniela Tinca, BSN, RN, ONC, Clinical Nurse, Department of Nursing, Hospital for Special Surgery, New York, NY.

Regina Cannon Drake, MA, RN, ONC, Patient Educator, Department of Nursing, Hospital for Special Surgery, New York, NY.

The authors have disclosed no conflicts of interest.

DOI: 10.1097/NOR.0000000000000382

et al., 2003; Konrath, Lock, & Goitz, 1996; Wahern, Torebjork, & Jorum, 1989). Cold therapy is used in sports and rehabilitation settings to treat tissue injuries and has been found to be effective in reducing pain and swelling after arthroscopy and arthroplasty surgery (Aide, Kwan, Naylor, Harris, & Mittal, 2012; Bleakley, et al., 2004; Kullenberg, Ylipaa, Soderlund, & Resch, 2006; Lessard, Scudds, Amedola, & Vaz, 1997). Cold therapy with com- pression has been shown to have additive benefits of re- duced swelling, but findings from the studies also sug- gest that gains derived from the combined treatment of cold and static compression are only marginally greater than either therapy alone (Block, 2010).

There are few studies specific to the use of cold ther- apy in the spine surgery patient population. In a rand- omized trial (N = 77) that measured analgesic effects of perioperative cold irrigation in patients who underwent lumbar microdiscectomy, researchers found the mean morphine equivalent dose used by patients who had cooling therapy to be significantly less (p < .001) than that used by patients who did not receive cooling ther- apy (Fountas et al., 1999). In a second experimental study (N = 30), researchers examined the effect of using a cooling pad in patients with unspecified lumbar sur- gery and noted that patients who received cold therapy

Copyright © 2017 by National Association of Orthopaedic Nurses. Unauthorized reproduction of this article is prohibited.

© 2017 by National Association of Orthopaedic Nurses Orthopaedic Nursing • September/October 2017 • Volume 36 • Number 5 345

reported significantly (p < .005) decreased narcotic con- sumption within the first 24 hours after surgery (Brander, Muro, Bromelly, & Hetreed, 1996). Researchers in a third experimental study of patients who had microdiscectomy (N = 36) compared the effect of a cooling system on postoperative pain after spine surgery and found no significant differences in pain scores between groups (Murata et al. 2014). In a fourth randomized trial (N = 41) of lumbar spine surgery pa- tients, investigators found patients who received post- operative therapy with a cooling device used 49% less patient-controlled analgesia (PCA) than the control group patients who did not receive cold therapy (Stephee, Booher, & Biscup, 1996). In addition, those who received cold treatment reported an eight out of 10 satisfaction score with the intervention.

Findings across the limited number of studies using cold therapies to manage the care of postoperative spine patients suggest benefits to decreasing narcotic con- sumption. Narcotics, specifically opiates, often require additional care management strategies to address ad- verse side effects including nausea, vomiting, constipa- tion, sedation, dizziness and the potential for tolerance, physical dependence, addiction, abuse, and diversion (Benyamin et al., 2008).

The purpose of this study was to determine the ef- fects of the application of cold (ice pack) on postopera- tive spine pain fusion patients and add to the body of knowledge specific to practical application in the spinal fusion surgical setting. Given limited clinical evidence, this study should inform practitioners on the benefits of using cold therapy as an adjuvant therapy to reduce pain and narcotic consumption in the management of the postoperative spine patient population.

Methods Design Researchers used a two-group randomized control de- sign to evaluate the effects of local cold therapy on post- operative pain and analgesia use after lumbar spinal fu- sion surgery. The primary outcome was postoperative pain. Secondary outcomes included analgesia use and perceived benefit of cold therapy. Participants were ran- domly assigned to two strata: (a) lumbar spine fusion patients with repositioning and application of cold ther- apy and (b) lumbar spine fusion patients with reposi- tioning and no application of cold therapy. Concealed randomization schedule was generated by the biostatis- tics department at the study organization, with rand- omization lists provided to investigators after consent and prior to treatment. Participants were assigned to either the control or intervention group by the random generator, and his/her treatment arm assignment was placed in a concealed envelope. There was no blinding of participants because it was obvious who received cold therapy and who did not. The study was approved by the organization’s institutional review board.

setting anD sample selection Subjects were recruited from July 11, 2013, through February 13, 2015, in their surgeons’ offices or the

preoperative holding center of an urban, acute care fa- cility specializing in the treatment of musculoskeletal diseases.

Participants were a convenience sample of adults who underwent postoperative lumbar spinal fusion sur- gery from one to four levels and were admitted to a spe- cific inpatient surgical unit within 12 hours after sur- gery. Preoperative inclusion criteria included (a) English speaking and (b) 18 years or older. Exclusion criteria included (a) non-English speaking, (b) intolerance to cold, (c) medical history of rheumatoid arthritis, sclero- derma, dermatological conditions, Reynaud’s phenom- ena, and dementia, and (c) use of ice to lower back prior to surgery.

Eligible participants were identified by the coinvesti- gating surgeons who communicated to the principal in- vestigator (PI), a nurse patient educator. The PI con- tacted potential participants during the preoperative scheduling process and served as the primary point per- son to discuss study details and introduce the consent form. The PI approached all identified potential partici- pants before surgery. The consent was reviewed, outlin- ing the approach and purpose of this study, which was to determine whether cold therapy in the form of local application of ice packs is effective in reducing postop- erative pain in patients who have spine fusion surgery. The concept of randomization was explained, inform- ing subjects that they would be chosen by chance, to give each participant an equal opportunity to be in- cluded in the ice versus no-ice study group provided they met study criteria postoperatively. Postoperative exclusion criteria were as follows: prolonged stay in the postanesthesia care unit, discharge to a nonstudy inpa- tient unit, and surgical intervention that changed dur- ing the operation.

Data on participant refusal were not collected for analysis. Ninety-three participants who gave study con- sent were not randomized to either intervention be- cause of failure to meet postoperative criteria.

measurement Pain levels were measured using the Numerical Rating Scale (NRS). This is a valid and reliable measurement of pain experienced by acute care patients (Hjermstad et al., 2011; Wewers & Lowe, 1990). Participants ver- bally reported a score from zero to 10, with zero repre- senting no pain and 10 representing the worst pain im- aginable. The NRS score was assessed both prior to and following study interventions, coincided with the staff protocol for routine 4-hour pain assessments and con- tinued for every 4 hours until discharge. Pain checks one through six were reported as the 24-hour group, and seven through 12 reported as the 48-hour group.

Analgesia use was obtained from the medication re- cord. Opioid use was converted to oral morphine equiv- alents with IV hydromorphone converted to IV mor- phine at a ratio of 2:10, and IV morphine converted to oral morphine at a ratio of 1:3 (Liu et al., 1995; Wu, 2005). Oral morphine conversion ratios were oxyco- done 1 mg: 1.5 mg oral morphine, propoxyphene 200 mg: 30 mg, hydrocodone 1:1, and hydromorphone 1:4 (Amabile & Bowman, 2006; Gordon et al., 1999; Wu, 2005).

Copyright © 2017 by National Association of Orthopaedic Nurses. Unauthorized reproduction of this article is prohibited.

346 Orthopaedic Nursing • September/October 2017 • Volume 36 • Number 5 © 2017 by National Association of Orthopaedic Nurses

Patient perception of benefit or evaluation of the in- tervention was a single item, yes or no question. Patients were asked “Did the intervention help to reduce your pain”?

proceDures

Intervention Group (Cold Therapy) During routine 4-hour pain assessment rounds, direct care registered nurses asked participants to rate their pain level using the NRS. Participants were turned and positioned to a side-lying position in bed. A Precept Premium Cold Pack sleeve was filled with 1 L of crushed ice and applied to the lower back, and the pack was sup- ported with a pillow. The cold packs were applied to the intervention group and a timer was set for 20 minutes. After 20 minutes the nurse removed the pack. Participants were again asked to rate their pain level using the NRS. Participants were also asked whether the intervention reduced their pain.

Control Group Procedures for the control group were the same as for the intervention groups, with the exception of the cold packs. During routine pain assessments, participants were asked to rate their pain level using the NRS. Participants were turned and positioned to a side-lying position in bed. After 20 minutes, the nurse asked the patient to rate his/her pain level using the NRS. Participants were also asked whether the intervention reduced their pain.

Nurses provided pain medication to all patients using a standardized pain medication order set. Data were re- corded every 4 hours and continued over a 48-hour pe- riod for both groups.

Data analysis It was determined that a sample size of 100 patients per group would provide 98% power at a two-sided α level of .05 for a two-sample t test to detect a two-point differ- ence in NRS pain score change between groups (assum- ing a within-group standard deviation of four points) at 24 and 48 hours postoperation. After an exploratory in- terim analysis postrandomization of 70 patients, the power analysis was revised. It was determined that a sample size of 68 patients per group would provide 80% power at a two-sided α level of .05 for a two-sample t test to detect a two-point difference in NRS pain score change between groups (assuming within-group stand- ard deviation of four points) at 24 hours postoperation. A total of 150 patients would be enrolled to account for attrition.

The cold therapy and control groups were compared for balance on baseline characteristics by calculating standardized differences. Standardized differences were calculated as the difference in means or mean rankings divided by the pooled standard deviation for normally distributed or ordinal variables and skewed continuous variables, respectively (Austin, 2009; Schacht, Bogaerts, Bluhmki, & Lesaffre, 2008). For cat- egorical variables, standardized differences were calcu- lated as the difference in proportions divided by the

pooled standard deviation (Austin, 2009). Imbalance was defined as a standardized difference with an abso- lute value greater than 21.96 0.322

74 =× (Austin, 2009).

Average change in the NRS pain score pre- to postin- tervention and cumulative PCA usage across pain checks one to six and seven to 12 were compared be- tween groups after adjustment for preoperative pain level using the generalized estimating equation (GEE) method with an identity link (Ma, Mazumdar, & Memtsoudis, 2012; Zeger, Liang & Albert, 1988). The GEE method accounts for the correlation between re- peated measurements on the same patient. Average change in the NRS pain score pre- to postintervention and cumulative PCA usage across pain checks one to 12 were compared between groups after adjustment for preoperative pain level using linear regression. Patient perception of intervention benefit was compared be- tween groups after adjustment for preoperative pain level using the GEE method with a logit link.

All statistical hypothesis tests were two-sided, with p < .05 considered statistically significant. Statistical analyses were performed with SAS Version 9.3 (SAS Institute, Cary, North Carolina).

Results stuDy participants One hundred and forty-eight patients completed the study. Participant characteristics are provided in Table 1. Data comparisons demonstrate characteristic similarity across groups. Data were reported on all completed pain checks at each interval. Because of patient discharge, withdrawal, or refusal, 63 patients in each group had complete data for the sixth pain check. The lowest num- ber of patients with complete data occurred during the 12th pain check with 53 and 61 patients in the cold ther- apy and control groups, respectively.

nrs score As summarized in Table 2 and Figure 1, our study showed the intervention group had a marginally greater reduction in mean NRS score across all 12 pain checks (M ± SD = −1.1 ± 0.8 points reduction vs. −1.0 ± 0.8 points reduction, p = .589). The triangles represent the means. The bottom and top of each rectangle indicate the first and third quartiles, respectively. The horizon- tal line within each rectangle indicates the median. The lines extending out of the bottom and top of each rectangle represent the minimum and maximum val- ues that lie within 1.5 times the interquartile range below and above the first and third quartiles, respec- tively. The results demonstrate both groups had re- duced pain but the reduction was not statistically or clinically significant.

analgesia use Findings specific to the second study aim are illustrated in Figure 2 boxplot. On average, the intervention group used less MSO4 mg equivalents (M ± SD = 12.6 ± 31.5 vs. 23.7 ± 40.0) than the control group across pain checks seven to 12 (p = .042).

Copyright © 2017 by National Association of Orthopaedic Nurses. Unauthorized reproduction of this article is prohibited.

© 2017 by National Association of Orthopaedic Nurses Orthopaedic Nursing • September/October 2017 • Volume 36 • Number 5 347

pain perception Results summarized in Table 2 noted that both groups reported a reduction in pain because of the intervention (nurse attention, i.e., turning and positioning) with or without ice.

Discussion In prior studies that examined the use of cold therapy in spine surgery patients, only one compared pain scores. These researchers did not find significant differences in

pain scores between comparison groups (Murata et al., 2014). Although these results are consistent with our study findings, it should be noted that the Murata study was small in size (N = 36) and participants were not randomized to treatment.

Our finding of significant reduction in analgesia use by the participants who received cold therapy is consist- ent with results of two similar studies that measured this outcome (Brander et al., 1996; Fountas et al., 1999). Significance of this finding cannot be overstated, as opi- oids are commonly used postoperatively by patients

table 1. participant characteristics

Control (n = 74) Cold Therapy (n = 74) Standardized Difference

Age, M ± SD 61.4 ± 14.9 62.4 ± 11.7 0.075

Female, n (%) 48 (64.9) 42 (56.8) −0.167

White, n (%) 64 (86.5) 66 (89.2) 0.083

Employment status, n (%)

Employed 42 (56.8) 44 (59.5) −0.047

Unemployed 5 (6.8) 8 (10.8) 0.121

Other 21 (28.4) 22 (29.7) −0.025

Unknown 6 (8.1) 0 (0) −0.420

Medical insurance, n (%)

Yes 70 (94.6) 74 (100) 0.239

Workman’s comp 2 (2.7) 0 (0) −0.239

Unknown 2 (2.7) 0 (0) −0.236

Duration of back pain, n (%) −0.174

<1 year 13 (17.6) 13 (17.6)

>1 year, <5 years 34 (45.9) 43 (58.1)

>5 years 26 (35.1) 18 (24.3)

Unknown 1 (1.4) 0 (0)

Current pain level, n (%) 0.088

No pain (NRS 0) 2 (2.7) 4 (5.4)

Mild (NRS 1–3) 8 (10.8) 3 (4.1)

Moderate (NRS 4–6) 26 (35.1) 27 (36.5)

Severe (NRS 7–10) 36 (48.6) 39 (52.7)

Unknown 2 (2.7) 1 (1.4)

Number of pain medication, n (%) −0.260

0 19 (25.7) 22 (29.7)

1 28 (37.8) 32 (43.2)

2 16 (21.6) 11 (14.9)

3 2 (2.7) 1 (1.4)

4 2 (2.7) 2 (2.7)

5 1 (1.4) 0 (0)

Unknown 6 (8.1) 6 (8.1)

Previous back surgery, n (%)

Yes 6 (8.1) 8 (10.8) 0.100

No 65 (87.8) 62 (83.8) −0.100

Unknown 3 (4.1) 4 (5.4) 0.064

Note. NRS = Numerical Rating Scale.

Copyright © 2017 by National Association of Orthopaedic Nurses. Unauthorized reproduction of this article is prohibited.

348 Orthopaedic Nursing • September/October 2017 • Volume 36 • Number 5 © 2017 by National Association of Orthopaedic Nurses

who have had spine surgery. Any effective adjuvant ther- apy that can mitigate the use of these narcotics is im- portant to consider, to lower the need for drug therapy and to avert side effects associated with opioids.

In our study, the similar perception of pain reduction between both groups raises interesting questions. First, is turning and positioning exclusively therapeutic be- cause both groups reported pain reduction after this in- tervention? Second, what is the effect of nurse attention on perceptions of postoperative pain? Additional study is needed to further understand the impact of reposi- tioning and time spent with nurses.

limitations Although the study was powered to establish the sample size, a larger sample is always preferable. The study was

conducted on one inpatient unit at an orthopaedic sur- gical specialty hospital; therefore, the results may not be generalized to other institutions.

implications for practice anD research The results demonstrate decreased analgesic consump- tion and a trend in lower NRS pain scores in the inter- vention group and a perceived reduction of pain in both groups regardless of intervention. The significant de- crease in narcotic consumption warrants a review of clinical practice to incorporate cold therapy in the standard postoperative order data set in this population. Nurse attention and the continuance of regular turning and positioning is also an important factor to continue, as it may positively influence patient perceptions related to pain. Independent clinical benefits associated with turning and positioning bear further investigation.

table 2. change in nrs, pca use, anD perceiveD pain reDuction

Control Cold Therapy Adjusted

Effect Size

n M ± SD n M ± SD Difference in

means [95% CI] df Chi-Square

Value p

Value

Change in pain

Pain checks 1-6 74 −1.0 ± 1.0 74 −1.1 ± 0.8 −0.1 [−0.4, 0.2] 1 0.20 .657

Pain checks 7-12 69 −1.2 ± 0.9 64 −1.1 ± 1.1 0 [−0.3, 0.3] 1 0 .986

Pain checks 1-12 74 −1.0 ± 0.8 74 −1.1 ± 0.8 −0.1 [−0.3, 0.2] 1 0.29 .589

Cumulative PCA usage

Pain checks 1-6 67 102.6 ± 91.5 69 90.0 ± 71.4 −15.4 [−43.6, 12.8] 1 1.15 .284

Pain checks 7-12 67 23.7 ± 40.0 63 12.6 ± 31.5 −13.6 [−26.7, −0.5] 1 4.12 .042

Pain checks 1-12 63 130.9 ± 119.8 59 105.2 ± 88.1 −29.3 [−66.1, 7.5] 1 2.41 .120

n n (%) n n (%) Odds Ratio [95% CI]

Perceived reduction in pain due to

intervention (yes/no)

Pain check 6 64 47 (73.4) 62 49 (79.0) 1.23 [0.55, 2.74] 1 0.26 .608

Pain check 12 60 47 (78.3) 53 46 (86.8) 1.62 [0.60, 4.34] 1 0.91 .340

Note. CI = confidence interval; NRS = Numerical Rating Scale; PCA = patient-controlled analgesia.

figure 1. Change in NRS scores across all 12 pain checks. NRS = Numerical Rating Scale. figure 2. MSO4 equivalents across all 12 pain checks.

Copyright © 2017 by National Association of Orthopaedic Nurses. Unauthorized reproduction of this article is prohibited.

© 2017 by National Association of Orthopaedic Nurses Orthopaedic Nursing • September/October 2017 • Volume 36 • Number 5 349

Summary The research described in this study was conducted to measure the effects of cold therapy on pain and analgesia use after spine fusion surgery. Although the results did not demonstrate a significant change in pain score, cold therapy was found to be effective in reducing analgesic pain medication use. This finding supports the use of cold therapy as a pain control alternative that can be used as an adjuvant therapy in spine surgery patients that can easily be incorporated into pain management schedules.

acknowleDgments The authors would like to acknowledge the 6 East Nursing staff, pharmacy, and study coinvestigators for their contributions and support: Frank P. Cammisa, MD; Matthew E. Cunningham, MD, PhD; Geraldine DiLorenzo, RN; James Farmer, MD; Federico Girardi, MD; Russel Charles Huang, MD; Alex P. Hughes, MD; Han Jo Kim, MD; Darren R. Lebl, MD; Joseph Nguyen, MPH; Michele Prigo, EdD; Bernard A. Rawlins, MD; Andrew A. Sama, MD; and Kelsey Vukov, RN.

references Abramson, D., Chu, L., & Tuck, S. (1966). Effect of tissue

temperatures and blood flow on motor nerve conduc- tion velocity. JAMA: The Journal of the American Medical Association, 198, 1082.

Adie, S., Kwan, A., Naylor, J., Harris, I., & Mittal, R. (2012). Cryotherapy following total knee replacement. Cochrane Database of Systematic Reviews, 9, 1–82.

Airaksinen, O. V., Kyrklund, N., Latvala, K., Kouri, J. P., Gronblad, M., & Kolari, P. (2003). Efficacy of cold gel for soft tissue injuries: A prospective randomized dou- ble-blinded trial. The American Journal of Sports Medicine, 31(5), 680–684.

Amabile, C., & Bowman, B. (2006). Overview of oral modi- fied-release opioid products for management of chronic pain. The Annals of Pharmacotherapy, 40, 1327–1329.

Austin, P. C. (2009). Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Statistics in Medicine, 28, 3083–3107.

Benyamin, R., Trescot, A., Datta, S., Buenaventura, R., Adlaka, R., Sehgal, N., … Vallejo, R. (2008). Opioid complications and side effects. Pain Physician Journal: Opioid Special Issue, 11, S105–S120.

Bleakley, C., McDonough, S., & MacAuley, D. (2004). The use of ice in the treatment of acute soft-tissue injury: A systematic review of randomized controlled trials. The American Journal of Sports Medicine, 32(1), 251–261.

Block, J. (2010). Cold and compression in the management of musculoskeletal injuries and orthopedic operative procedures: A narrative review. The Journal of Sports Medicine, 1, 105–113.

Brander, B., Munro, B., Bromely, L., & Hetreed, M. (1996). Evaluation of the contribution to postoperative anal- gesia by local cooling of the wound. Anaesthesia, 51, 1021–1025.

Fountas, K., Eftychia, K., Johnston, K., Smission, H., Vogel, R., & Robinson, J. (1999). Postoperative lumbar micro discectomy pain. Spine, 24, 1958–1960.

Gordon, D. B., Stevenson, K. K., Griffie, J., Muchka, S., Rapp, C., & Ford-Roberts, K. (1999). Opioid equianal- gesic calculations. Journal of Palliative Medincine, 2, 209–219.

Hjermstad, M. J., Fayers, P. M., Haugen, D. F., Caraceni, A., Hanks, G. W., & Loge, J. H. (2011). European Palliative Care Research Collaborative (EPCRC). Studies com- paring numerical rating scales, verbal rating scales, and visual analogue scales for assessment of pain in- tensity in adults: A systematic literature review. Journal of Pain and Symptom Management, 41(6), 1073–1093. doi:10.1016/j.jpainsymman.2010.08.016

Konrath, G., Lock, T., & Goitz, H. (1996). The use of cold therapy after anterior cruciate ligament reconstruc- tion: A prospective randomized study and literature review. American Journal of Sports Medicine, 24, 629–633.

Kullenberg, B., Ylipaa, S., Soderlund, K., & Resch, S. (2006). Postoperative cryotherapy after total knee ar- throplasty: A prospective study of 86 patients. The Journal of Arthroplasty, 21(8), 1175–1179. doi:10.1016/j. arth.2006.02.159

Lessard, L., Scudds, R., Amendola, A., & Vaz, M. (1997). The efficacy of cryotherapy following arthroscopic knee surgery. Journal of Orthopaedic & Sports Physical Therapy, 26(1), 14–22.

Levy, A., & Marmar, E. (1993). The role of cold compres- sion dressing in the postoperative treatment of total knee arthroplasty. Clinical Orthopaedics, 297, 174–178.

Liu, S. S., Carpenter, R. L., Mulroy, M. F., Weissman, R. M., McGill, T. J., Rupp, S. M., & Allen, H. W. (1995). Intravenous versus epidural administration of hydro- morphone: Effects on analgesia and recovery after radical retropubic prostatectomy. Anesthesiology, 82(3), 682–688.

Ma, Y., Mazumdar, M., & Memtsoudis, S. G. (2012). Beyond repeated-measures analysis of variance: Advanced sta- tistical methods for the analysis of longitudinal data in anesthesia research. Regional Anesthesia and Pain Medicine, 37, 99–105.

Murata, K., Yoshimoto, M., Takebayashi, T., Ida, K., Nakano, K., & Yamashita, T. (2014). Effect of cryother- apy after spine surgery. Asian Spine Journal, 8(6), 753– 758. doi:10.4184/asj.2014.8.6.753

Schacht, A., Bogaerts, K., Bluhmki, E., & Lesaffre, E. (2008). A new nonparametric approach for baseline covariate adjustment for two-group comparative stud- ies. Biometrics, 4, 1110–1116.

Stefee, A., Booher, J., & Biscup, R. (1996). Evaluation of cold therapy in postoperative spine patients. Surgical Technology, 5, 385–387.

Wahern, L., Torebjork, E., & Jorum, E. (1989). Central sup- pression of cold-induced C-fiber pain by fiber myeli- nated fiber input. Pain, 38, 313–319.

Wewers, M. E., & Lowe, N. K. (1990). A critical review of visual analogue scales in the measurement of clinical phenomena. Research in Nursing & Health, 13(4), 227– 236.

Wu, C. (2005). Acute postoperative pain. In R. D. Miller, (Ed.): Miller’s anesthesia (6th ed., pp. 2745–2747). Philadelphia, PA: Elsevier Churchill Livingstone.

Zeger, S., Liang, K., & Albert, P. (1988). Models for longitu- dinal data: A generalized estimating equation ap- proach. Biometrics, 44, 1049–1060.