RESEARCH

1 57
2 58
3
4
Effects of Connective Tissue Massage 59
60
5
6 on Pain in Primiparous Women After 61
62
7 63
8
9
Cesarean Birth 64
65
10 Mary Colleen Simonelli, Louise T. Doyle, MaryAnn Columbia, Phoebe D. Wells, Kelly V. Benson, Q13 66
11 and Christopher S. Lee 67
12 68
13 69
14 Correspondence ABSTRACT 70
15 Mary Colleen Simonelli, 71
RN, PhD, Boston College Objective: To evaluate the efficacy of connective tissue massage to reduce postoperative pain in primiparous women
16 72
Connell School of Nursing, on Postoperative Day 1 after unplanned cesarean birth.
17 140 Commonwealth Ave., 73
Design: A randomized controlled trial with three groups: intervention (Group 1), control or standard care (Group 2),
18 Maloney Hall 239, Chestnut 74
and individualized attention (Group 3).
19 Hill, MA 02467. 75
20 mary.simonelli@bc.edu Setting: Family/newborn units of a large teaching hospital in the Northeastern United States. 76
21 Keywords Participants: A total of 165 women who experienced unplanned cesarean births of singleton newborns at term Q1 77
22 complementary therapies gestation. 78
23 massage 79
Methods: Participants were randomized to three groups: those in Group 1 received a 20-minute massage, those in
opioids
24 Group 2 received the usual standard of care, and those in Group 3 received 20 minutes of individualized attention. On 80
pain
25 relaxation Postoperative Day 1, participants completed questionnaires to measure overall pain, stress, and relaxation at Time 1 81
26 stress and again 60 minutes later. Daily numeric pain ratings and medication consumption data were retrieved from the 82
27 unplanned cesarean birth electronic health care records. Latent growth modeling and analysis of variance were used to analyze data, as 83
28 appropriate. 84
29 Results: Participants in Group 1 had increased relaxation (p < .001), decreased pain (p < .001), decreased stress Q2 85
30 (p < .001), and decreased opioid use on Day 1 (p ¼ .031) and Day 2 (p ¼ .006) of the hospital stay after the inter- 86
31 vention compared with the other groups. Additionally, opioid use in Group 1 decreased linearly, whereas the control 87
32 groups had a nonlinear pattern of change. 88
33 Conclusion: Using massage therapy during postoperative hospitalization improved relaxation and decreased pain, 89
34 stress, and opioid use in this sample of women after unplanned cesarean births. 90
35 91
JOGNN, -, -–-; 2018. https://doi.org/10.1016/j.jogn.2018.07.006
36 92
Accepted July 2018
37 93
38 94
39 95
40 96
41 97
P ostoperative pain management for women the management of postoperative pain and stress
Mary Colleen Simonelli,
42 RN, PhD, is the Assistant who have given birth through unplanned may play an important role in routine clinical 98
Department Chair and a
43 cesareans requires a multimodal, integrative practice after cesarean birth. 99
clinical professor in the
44 William F. Connell School approach. The traditional practice of adminis- 100
45 of Nursing, Boston College, tering opioids for pain control impairs women’s There have been only a few studies on non- 101
46 Chestnut Hill, MA, and a
abilities to remain awake and alert during the pharmacologic interventions for post-cesarean 102
staff member, Newborn
47 Family Unit, Massachusetts critical transition to parenting in general and to birth pain, and although evidence was found for 103
48 General Hospital, Boston, bond with, breastfeed, and nurture their new- the effectiveness of massage as adjuvant therapy 104
49 MA. borns in particular (Sakalis, William, Hepworth, for post-cesarean pain, the samples in the 105
50 Hartmann, & Tamimi, 2013; Zanardo et al., studies comprised women who had elective ce- 106
51 2010). In addition to postoperative pain, un- sarean births (Abbaspoor, Akbari, & Najar, 2014; 107
(Continued)
52 planned cesarean birth is associated with psy- Saatsaz, Rezaei, Alipour, & Beheshti, 2016). 108
53 chological stress that further complicates Moreover, there is a paucity of empirical evidence 109
54 maternal self-care and care of the newborn to support any specific clinical practices that can 110
The authors report no con-
55 flict of interest or relevant (Modarres, Afrasiabi, Rahnama, & Montazen, be used to reduce psychological stress after 111
56 financial relationships. 2012). Hence, nonpharmacologic strategies for unplanned cesarean. Although previous 112

http://jognn.org ª 2018 AWHONN, the Association of Women’s Health, Obstetric and Neonatal Nurses. 1
Published by Elsevier Inc. All rights reserved.
FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce
 RESEARCH Connective Tissue Massage After Cesarean Birth

113 beyond those associated with elective cesar- 169

114 Women who experienced unplanned cesarean births have eans, it is important to include them in research 170
115 been excluded from prior research on optimal pain regarding postoperative pain management. 171
116 management techniques. 172
117 173
118
Role of the Maternity Care Provider 174
Q3 researchers identified the unique challenges Perinatal nurses provide physical care, emotional
119 175
faced by women who had unplanned cesareans, support, and education to new mothers and
120 176
this group of women has largely been excluded families. Critical to this role is the reduction of pain
121 177
from prior research on interventions to improve and stress to promote wellness for the mother
122 178
maternal outcomes. To address these important and her newborn (Abbaspoor et al., 2014).
123 179
knowledge gaps, the purpose of this prospective Unplanned cesarean birth is associated with an
124 180
randomized controlled trial was to test the effi- increase in pain and stress compared with
125 181
cacy of massage therapy as a relaxation, pain, vaginal birth. Surgical incisions and exhaustion
126 182
and stress management strategy in primiparous after labor and surgery can impair a mother’s
127 183
women who experienced unplanned cesareans. ability to bond with, breastfeed, and nurture her
128 184
We hypothesized that the use of massage ther- newborn (Zanardo et al., 2010). In prior studies,
129 185
apy would improve relaxation and reduce pain, researchers suggested that emergent or un-
130 186
stress, and medication use in participants planned cesarean births could have negative ef-
131 187
assigned to this intervention group compared fects on breastfeeding, especially during the
132 188
with standard of care and individualized attention initial postpartum period (Sakalis et al., 2013;
133 189
groups. Zanardo et al., 2010).
134 190
135 191
136 During the past two decades, cesarean rates Massage Therapy 192
137 have increased dramatically worldwide (Betrán Massage therapy may present an effective way to 193
138 et al., 2016). Despite a lack of evidence to link manage postoperative pain and decrease opioid 194
139 cesarean birth with improved maternal and use. Connective tissue massage can decrease 195
140 neonatal outcomes, currently in the United States pain perception and increase b-endorphin pro- 196
141 nearly one third of all newborns are born by ce- duction (Bauer et al., 2010). Researchers at the 197
142 sarean (Martin, Hamilton, Osterman, Driscoll, & Buck Institute for Research on Aging at McMaster 198
Louise T. Doyle, RNC-OB,
143 Drake, 2018). Given this reality, nurses must University in Ontario, Canada, showed that mas- 199
HNB-BC, LMT, is a
144 resource/staff nurse, evaluate current practices and develop effective sage works on a cellular level to decrease 200
145 Newborn Family Unit, pain management strategies that do not inflammation and increase the energy-producing 201
Massachusetts General compromise a woman’s ability to bond with or
146 organelles of the muscle (Turchaninov, 2011). 202
Hospital, Boston, MA.
147 care for her newborn effectively. McRee et al. (2007) found the acute response to 203
148 MaryAnn Columbia, RN, 204
massage was equal to a dose of morphine in
IBCLC, is a lactation
149 consultant/resource/staff Women who experience cesarean births are at improving sleep and decreasing anxiety for 205
150 nurse, Newborn Family risk for anesthetic, surgical, and postoperative women after gynecologic surgery. 206
151 Unit, Massachusetts General complications (Clark & Silver, 2011). They have 207
Hospital, Boston, MA.
152 longer hospitalizations and recoveries after Authors of prior clinical studies showed the pos- 208
153 Phoebe D. Wells, RN, MS, childbirth compared with women who experience itive effects of back massage on the reduction of 209
154 HWNC-BC, is a lactation 210
vaginal births. In our institution, the length of a pain and anxiety after cardiac and thoracic sur-
consultant/resource/staff
155 nurse, Newborn Family typical hospitalization after a cesarean is 96 gery (Adams, White, & Beckett, 2010; Bauer 211
156 Unit, Massachusetts General hours, whereas the standard hospital stay after a et al., 2010; Cutshall et al., 2010). Additionally, 212
157 Hospital, Boston, MA. vaginal birth is 48 hours. The current standard of massage decreased anxiety and muscular ten- 213
158 Kelly V. Benson, RN, LMT, care at our institution after cesarean is to admin- sion and increased relaxation and patient satis- 214
159 is staff nurse, Newborn ister long-acting intrathecal morphine along with faction compared with standard care after 215
160 Family Unit, Massachusetts 216
the anesthetic for surgery, timed nonsteroidal cardiac surgery (Bauer et al., 2010). Dion et al.
General Hospital, Boston,
161 MA. anti-inflammatory drugs (NSAIDs) for the first 24 (2011) reported that massage, in conjunction 217
162 hours after surgery, and oral opioids and NSAIDs with regular pain medication, significantly 218
Christopher S. Lee, PhD,
163 RN, FAHA, FAAN, as needed for the duration of the woman’s hos- improved patient pain and anxiety after major 219
164 FHFSA, is the Associate pitalization. At discharge, women who experi- surgery. 220
165 Dean of Research and enced cesarean births receive a prescription for 221
166 professor, William F. 222
opioids. Because women who experience un- In their meta-analysis, Boyd et al. (2016) identi-
Connell School of Nursing,
167 Boston College, Chestnut planned cesareans are at increased risk for fied the following issues with the 16 studies (12 223
168 Hill, MA. physical and psychological complications high quality and 4 low quality) on massage 224

2 JOGNN, -, -–-; 2018. https://doi.org/10.1016/j.jogn.2018.07.006 http://jognn.org

FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce

 Simonelli, M. C. et al. RESEARCH

225 therapy included in their review: (a) the dosage 281

226 of massage therapy has not been examined; (b) Massage therapy decreased pain and stress and improved 282
227 researchers have not used a no-treatment con- relaxation for primiparous women who experienced 283
228 trol group, nor have they accounted for the unplanned cesarean births. 284
229 unintentional effects of individualized attention 285
230 and touch by study staff; (c) although pain and 286
pain in women after cesareans (Abbaspoor et al.,
231 anxiety have been measured, effects on opioid 287
2014; Saatsaz et al., 2016). The nurse who cares
232 use have not been reported; and (d) there has 288
for a woman after an unplanned cesarean birth is
233 been no evaluation of massage therapy on 289
her primary pain manager and is responsible for
234 length of stay after surgery. Many studies took 290
helping the woman with pain relief. Excellent
235 place outside the United States. Because 291
clinical practice incorporates dynamic, multi-
236 80% of the world’s opioids are consumed in the 292
modal pain management strategies and pro-
237 United States, and women ages 20 to 44 years 293
motes a healing environment. Researchers
238 consume 30% more opioids than men (Express 294
showed the clinical effectiveness of many com-
239 Scripts, 2014; Jones, 2013), we believed it was 295
plementary and alternative medicine in-
240 imperative to examine the effectiveness of 296
terventions, and in recent studies, massage was
241 alternative therapies, such as massage, not only 297
found to be cost effective (Abbaspoor et al.,
242 on pain itself, but also on opioid use in U.S. 298
2014; Saatsaz et al., 2016). The use of massage
243 women after cesareans. 299
as an integrative postoperative pain intervention
244 300
may reduce pain and stress and increase mobility
245 301
246 Massage Therapy in Childbearing and relaxation; consequently, it may reduce infant
302
feeding issues and promote satisfaction with
247 Women 303
hospitalization for women after unplanned cesar-
248 Despite a growing body of research about mas- 304
ean births.
249 sage therapy as an effective intervention for pain 305
250 management, very few studies have been con- 306
251 ducted with women during pregnancy Conceptual Framework 307
252 (Chuntharapat, Petpichetchian, & Hatthakit, We used Kolcaba’s comfort theory as the con- 308
253 Q4 2008), during labor (Dehcheshmeh & Rafiei, ceptual framework for this research. Kolcaba 309
254 2015; Taghinejad, Delpisheh, & Suhrabi, 2010), (1992, 1994) directs nurses to assess all the 310
255 or in the postpartum period (Abbaspoor et al., health care needs of a patient, including the 311
256 2014; Hanan, Kamilla, Ahmed, & Amina, 2014; factors that cannot be changed, and to design 312
257 Saatsaz et al., 2016; Xue et al., 2016). In a small holistic interventions to meet those needs by 313
258 Turkish study, Abbaspoor et al. found that hand measuring the outcome of the intervention 314
259 and foot massage after elective cesarean low- compared with the pre-intervention baseline. 315
260 ered pain and anxiety. In a 2016 single-blind trial With Kolcaba’s comfort theory, we acknowledge 316
261 of 156 primiparous Iranian women who experi- that enhanced comfort is achieved if 317
262 enced elective cesareans, researchers random- interventions provide relief, ease suffering, and 318
263 ized participants to a 10-minute foot massage, a allow transcendence. Although analgesics pro- 319
264 20-minute hand and foot massage, or a control vide relief, the inclusion of further interventions to 320
265 group of usual care. They found a significant address stress and anxiety will ease suffering. 321
266 reduction in pain and anxiety (p < .001) and an Ultimately, the goal is to help the patient to rise 322
267 increase in breastfeeding frequency after either above the challenges brought on by health care 323
268 massage intervention. However, there was no needs. Researchers have used the comfort the- 324
269 significant difference between the two interven- ory to conduct studies with pregnant women and 325
270 tion groups (Saatsaz et al., 2016). women after cesarean. Chuntharapat et al. 326
271 (2008) found that women randomized to a yoga 327
272 The United States is currently experiencing an program in the prenatal period had greater 328
273 opioid crisis; consequently, health care providers levels of comfort, decreased perceived labor 329
274 must explore alternative and complementary pain, and a shorter duration of labor than women 330
275 methods of pain management for postsurgical in a control group. In a study of women who 331
276 patients, particularly for those with unplanned experienced cesareans, participants in the 332
277 surgery. Massage decreases the responsiveness experimental group who received care based on 333
278 of pain receptors and increases b-endorphin the comfort theory had greater comfort levels 334
279 production and therefore may serve as an effec- than those in the control group (Derya & 335
280 tive supportive or primary intervention to reduce Pasinliog lu, 2017). 336

JOGNN 2018; Vol. -, Issue - 3

FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce

 RESEARCH Connective Tissue Massage After Cesarean Birth

337 Methods histories of previous cesareans, had multiple 393

338 Design gestations, planned to have cesarean birth for 394
339 We conducted a three-group randomized any reason, or had infants in the NICU were also 395
340 controlled trial to evaluate the efficacy of mas- excluded. As an incentive, participants were 396
341 sage on post-cesarean birth pain, stress, and entered in a drawing for one of five $50 gift cards 397
342 relaxation. Participants were randomized to the as part of their participation in the study. 398
343 20-minute massage intervention group, standard 399
344 Instruments 400
care control group, or individualized attention
345 All participants completed a visual analog scale 401
group. The hospital institutional review board
346 (VAS) to describe baseline pain, stress, and 402
approved the study. Study staff not engaged in
347 relaxation levels. The VAS contains numeric and 403
direct patient care at the time obtained written
348 written indicators of pain intensity and levels of 404
informed consent from each participant, and all
349 stress and relaxation. Bijur, Silver, and Gallagher 405
participants were informed of risks, benefits, and
350 (2001) assessed the reliability of use of the VAS 406
their right to withdraw consent at any time. All
351 to measure acute pain in emergency department 407
study materials were kept in a locked box located
352 patients. They reported a 0.97 summary intra- 408
in the locked nursery of the locked inpatient
353 class correlation coefficient for paired VAS 409
postpartum unit. Data entered into the database
354 scores, and Bland Altman analysis showed that 410
were stored on the principle investigator’s
355 90% of the ratings were within 9 mm of each other 411
password-protected, encrypted laptop.
356 (Bijur, Silver, & Gallagher, 2001). We asked par- 412
357 ticipants to record pain, stress, and relaxation 413
358 Setting and Sample levels on paper copies of the VAS to maintain 414
359 The study was conducted at a large teaching consistency with previous studies (Adams, White, 415
360 hospital in the Northeastern United States. There & Beckett, 2010; Bauer et al., 2010; Cutshall 416
361 are approximately 3,500 births per year in this et al., 2010; Dion et al., 2011; Saatsaz, et al., 417
362 institution, and its primary cesarean birth rate is 2016; Wentworth et al., 2009). Participants 418
363 20.2%. The sample comprised primiparous circled the number, from 0 to 10, that corre- 419
364 women who experienced unplanned cesarean sponded with the intensity of their pain and 420
365 births at term gestation. From an initial group of perceived stress and relaxation levels. Paper 421
366 60 participants, we calculated effect sizes and surveys to collect demographic information; Q6 422
367 estimated the full sample size. We needed a overall perceptions of pain, stress, relaxation, and 423
368 sample size of 53 participants per group to detect newborn feeding; and satisfaction with hospital 424
369 a difference in pain, stress, and relaxation among stay were distributed to all participants at the time 425
370 the massage, standard of care, and individual- of enrollment. 426
371 ized attention groups. We enrolled 55 participants 427
372 per group (N ¼ 165) to account for potential Average daily pain score and opioid and NSAID 428
373 attrition and/or incomplete data. medication use of the participants were deter- 429
374 mined from review of medication administration 430
375 Women were eligible to participate if they were records in the electronic health care record. To 431
376 primiparous and had experienced unplanned calculate the average daily pain score, we added 432
377 cesarean births. Participants were recruited on every pain assessment recorded in a 24-hour 433
378 Postoperative Day 0 or early Postoperative Day 1 period after the birth time and divided it by the 434
379 during their postpartum stay. The principle number of assessments. To calculate daily opioid Q7 435
380 investigator or trained study nurses screened for and NSAID use, we added the amount of each 436
381 eligibility and obtained informed consent. We medication given in milligrams in a 24-hour period 437
382 excluded women from participation if they had after the birth time. Oxycodone, in 5-mg or 10-mg 438
383 preexisting histories of significant psychological doses, is the standard medication ordered for 439
384 problems; were not able to understand written women after cesarean births at our institution. 440
385 and verbal instructions; had received general When participants with orders for Dilaudid (Pur- 441
386 anesthesia or had used patient-controlled anal- due Pharma, Stamford, CT) in 2-mg or 4-mg Q8 442
387 gesia; had damaged skin tissue, arthritis, phle- doses were enrolled in the study, we converted 443
388 bitis, burn wounds in the area to be massaged, the dosage of that medication for transcription in 444
389 injury, inflammation, eczema, significant history of the database to the equivalent 5 mg or 10 mg of 445
390 cardiovascular issues, or diagnosis of a respira- oxycodone. This maintained consistency of the 446
391 Q5 tory disease; or did not feel comfortable receiving dosages across participants. Because body 447
392 a massage or being touched. Women who had mass index (BMI) affects medication absorption 448

4 JOGNN, -, -–-; 2018. https://doi.org/10.1016/j.jogn.2018.07.006 http://jognn.org

FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce

 Simonelli, M. C. et al. RESEARCH

449 (Ingrande & Lemmens, 2010), it was important to hospitalization. Study staff collected these data 505
450 evaluate opioid use in conjunction with BMI. from the electronic charting system to remain 506
451 Consequently, we calculated daily and total blinded to study group assignment. 507
452 opioid/BMI and total NSAID/BMI by adding daily 508
453 opioid and NSAID use in milligrams and dividing Analysis 509
454 the total by the participant’s pre-pregnancy BMI. Data were entered into a Microsoft Excel spread- 510
455 With these data, we were able to determine the sheet and were analyzed with the use of SPSS for 511
456 group differences for pain, stress, and relaxation Windows version 24.0. Descriptive statistics were 512
457 to evaluate the average daily pain level and used to describe the sample and evaluate statis- 513
458 opioid and NSAID use for each participant. tical assumptions. Demographics of the study 514
459 participants were analyzed to assess the 515
460 Procedures randomization of those variables among the 516
461 We used a sealed-envelope technique to groups. Latent growth modeling was used to 517
462 randomize participants to massage (Group 1), characterize and compare change in pain and 518
463 standard care (Group 2), or individualized attention related outcomes over time. Characterization en- 519
464 (Group 3). Participants randomized to Group 2 had tails the quantification of intercepts (i.e., pre- 520
465 no changes in their care. Participants randomized intervention values) and slopes (i.e., rate of 521
466 to Group 3 spent 20 minutes with the study staff change over time); data are reported in raw 522
467 discussing their birth experiences. Because dis- values, standard errors (SE), values/SE (t statis- 523
468 tractions are known to affect patients’ perceptions tic), and two-sided p values. Comparisons among 524
469 of pain, the intervention of individualized attention the three groups were made using random effects 525
470 was included to avoid attribution of the potential among intercepts and slope (taking into account 526
471 effects of the massage to receptive attention rather the influence of the intervention groups on the in- 527
472 than to the massage itself (Kohl, Winfried, & tercepts and slopes simultaneously); results are 528
473 Glombiewski, 2013; Sprenger et al., 2012). Partic- reported as t statistics and two-sided p values. 529
474 ipants in Group 1 received massage from the co- Latent growth modeling was completed with the 530
475 investigator, who is a licensed massage therapist use of Mplus version 8 (Malacca Securities, n.d.). Q9 531
476 with certification in perinatal massage and board 532
477 certification in holistic nursing, or by a study nurse 533
478 who also is licensed as a massage therapist. Results 534
479 Demographic Variables 535
480 Initially, the study staff placed pieces of paper Demographic and clinical characteristics of the 536
481 with the group assignments into opaque enve- three groups are presented in Table 1. The mean 537
482 lopes. The envelopes were sealed, shuffled, and age of the participants was 32.5 years. Most were 538
483 stacked to ensure randomization of group married (72.9%), White (70.5%), and non- 539
484 assignment. The top envelope was brought to the Hispanic (80.2%). Their mean pre-pregnancy 540
485 participant once she gave informed consent. BMI was 27.13 kg/m2. There were no significant 541
486 Each participant was asked to open the sealed differences among groups for any of the de- 542
487 envelope. Inside, each found a slip of paper with mographic or clinical characteristics. 543
488 the words Group 1, Group 2, or Group 3 printed 544
489 on it to designate group assignment. Pain (Primary Efficacy Outcome) 545
490 We found no differences among the groups in 546
491 On Postoperative Day 1 (24–48 hours after the pain measured by VAS at baseline. Pain 547
492 birth), a portable monitor was used to record improved significantly between Time 1 and Time 548
493 blood pressure and pulse measurements, and all 2 in the massage group (–1.8  0.2, t ¼ –7.9, p < 549
494 participants completed the baseline VAS for pain, .001) and individualized attention group (–0.4  550
495 stress, and relaxation (Time 1). Sixty minutes after 0.2, t ¼ –2.0, p ¼ .046) but not in the standard of 551
496 VAS and vital sign assessment for the standard care group (p ¼ .874). Improvement in pain over 552
497 care or 60 minutes after the initiation of the mas- time differed significantly among the three groups 553
498 sage or individualized attention, participants in favor of the massage intervention (t ¼ 4.2, p < 554
499 completed the VAS a second time and vital sign .001; see Figure 1). However, we found no sta- 555
500 measurements were repeated (Time 2). Nursing tistically significant differences in average daily 556
501 staff use a 0-to-10 verbal, numeric pain rating pain scores documented in the health care re- 557
502 scale routinely (per hospital protocol), and these cord among the groups at baseline (t ¼ 0.6, p ¼ 558
503 results were used to derive average daily pain .589) or over the course of the hospitalization (t ¼ 559
504 ratings during each 24-hour period of the 0.8, p ¼ .427; see Figure 2). 560

JOGNN 2018; Vol. -, Issue - 5

FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce

 RESEARCH Connective Tissue Massage After Cesarean Birth

561 617
562 Table 1: Demographics and Physical Characteristics by Study Group 618
563 619
Characteristic Massage (n ¼ 55) Standard Care (n ¼ 55) Individualized Attention (n ¼ 55) p
564 620
Race, n (%)
565 621
566 Asian 1 (1.8) 5 (9.1) 9 (16.4) .085 Q11 622
567 Black 3 (5.5) 5 (9.1) 4 (7.3) 623
568 624
White 44 (80.0) 38 (69.1) 35 (63.6)
569 625
570 Other 4 (7.3) 3 (5.5) 6 (10.9) 626
571 Unknown 2 (3.6) 4 (7.3) 1 (1.8) 627
572 628
Native American 1 (1.8) 0 (0) 0 (0)
573 629
574 Ethnicity, n (%) 630
575 Hispanic 5 (9.1) 3 (5.5) 9 (16.4) .738 631
576 632
Non-Hispanic 45 (81.8) 44 (80) 45 (81.8)
577 633
578 Unknown 5 (9.1) 8 (14.5) 1 (1.8) 634
579 Marital status, n (%) 635
580 Married 37 (67.3) 40 (72.7) 44 (80.0) .149
636
581 637
Single 15 (27.3) 15 (27.3) 11 (20.0)
582 638
583 Divorced 3 (5.5) 0 (0) 0 (0) 639
584 Age, years 640
585 641
Range 17–46 19–40 21–44 .319
586 642
587 Mean 32.25 31.96 33.25 643
588 SD 5.049 4.451 4.522 644
589 645
Median 32.00 32.00 33.00
590 646
591 Body mass index, kg/m2 647
592 Range 19.80–45.35 18.00–40.47 18.90–55.5 .95 648
593 649
Mean 26.9418 27.2960 27.1538
594 650
SD 5.33452 5.44352 6.63809
595 651
596 Median 26.2100 26.6700 25.8500 652
597 653
Note. SD ¼ standard deviation.
598 654
599 655
600 656
601 657
602 Stress Relaxation 658
603 Stress scores improved significantly between Relaxation improved significantly from Time 1 to 659
604 Time 1 and Time 2 in the massage group (–2.3  Time 2 in the massage group (–3.1  0.3, t ¼ 660
605 0.2, t ¼ –9.4, p < .001) and individualized atten- –9.8, p < .001) but not in the individualized 661
606 tion group (–0.3  0.1, t ¼ –2.5, p ¼ .014) but not attention group (p ¼ .146) or standard of care 662
607 in the standard of care group (p ¼ .918; see group (p ¼ .718; see Figure 4). There was less 663
608 Figure 3). There was more stress at baseline relaxation at baseline among participants in the 664
609 among participants in the massage group than massage group compared with those in the other 665
610 participants in the other groups (t ¼ 4.4, p < study groups (t ¼ 2.9, p ¼ .003); even when this 666
611 .001); even when this was considered, improve- was considered, improvement in relaxation 667
612 ment in stress differed significantly over time differed significantly over time among the three 668
613 among the three groups, favoring the massage groups, favoring the massage intervention (t ¼ 669
614 intervention (t ¼ 7.2, p < .001). 5.8, p < .001). 670
615 671
616 672

6 JOGNN, -, -–-; 2018. https://doi.org/10.1016/j.jogn.2018.07.006 http://jognn.org

FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce

 Simonelli, M. C. et al. RESEARCH

673 729
674 730
675 731
676 732
677 733
678 734
679 735
680 736
681 737
682 738
683 739
684 740
685 741
686 742
687 743
688 744
689 745
690 746
691 747
692 748
693 749
Figure 1. Pain before (Pre) and after (Post) the intervention. Change in pain was significant in response to the massage
694 750
(intercept [i] ¼ 3.94  0.24, t ¼ 16.16, p < .001 and slope [s] ¼ –1.80  0.23, t ¼ –7.87, p < .001) and individualized attention
695 751
interventions (i ¼ 3.52  0.26, t ¼ 13.47, p < .001 and s ¼ –0.42  0.21, t ¼ –1.99, p ¼ .046) but not to standard care (i ¼ 3.78
696 752
 0.20, t ¼ 18.03, p < .001 and s ¼ –0.03  0.17, t ¼ –0.16, p ¼ .874). Improvement in pain over time differed significantly
697 753
among the three arms, favoring the massage intervention (t ¼ 4.2, p < .001).
698 754
699 755
700 756
701 757
702 758
703 759
704 760
705 761
706 762
707 763
708 764
709 765
710 766
711 767
712 768
713 769
714 770
715 771
716 772
717 773
718 774
719 775
720 776
721 777
Figure 2. Average pain scores at each postoperative day of hospitalization. Although there were no differences in the
722 778
improvement of average daily pain among the three groups (t ¼ 0.8, p ¼ .427), average daily pain improved significantly over
723 779
time in the massage intervention group (slope [s] ¼ –0.23  0.06, t ¼ –3.63, p < .001), standard care group (s ¼ –0.12  0.06,
724 780
t ¼ –2.02, p ¼ .043), and the individualized attention group (s ¼ –0.17  0.06, t ¼ –2.76, p ¼ .006). AVE ¼ average daily pain
725 781
from the electronic medical record; POD ¼ postoperative day.
726 782
727 783
728 784

JOGNN 2018; Vol. -, Issue - 7

FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce

 RESEARCH Connective Tissue Massage After Cesarean Birth

785 Discussion 841

786 A multimodal approach to pain management is necessary 842
Our results provide preliminary evidence for the
787 for women who must remain alert and care for their 843
use of connective tissue massage, because they
788 newborns after unplanned cesarean births. 844
suggest that it decreases pain and stress and
789 promotes relaxation in women who have experi- 845
790 enced unplanned cesarean births. Additionally, 846
791 Opioid and NSAID Use 847
There were fewer requests for opioids on Post- participants who received 20-minute massages on
792 Postoperative Day 1 not only consumed fewer 848
793 operative Days 1 and 2 among participants in the 849
massage group than among those in the other opioids in the 24-hour period that included the
794 massage but consumed fewer opioids on Day 2, 850
795 study groups (t ¼ 2.2; p ¼ .031 and p ¼ .006, 851
respectively). Opioid use decreased significantly the 24-hour period after the intervention. At the
796 institution where the study took place, standard 852
797 over time in all groups (see Figure 5). When 853
opioid use from Day 1 until discharge was practice includes provider evaluation of the
798 amount of opioids consumed during hospitaliza- 854
799 compared, we found no statistical difference in 855
the reduction of opioid use among the three tion and discussion of this information with women
800 to individualize the amount of opioids prescribed at 856
801 groups (t ¼ 0.6, p ¼ .559). 857
discharge. Although it is difficult to determine if the
802 difference we found for Days 1 and 2 had an effect 858
We found no significant differences in NSAID use
803 on the opioids prescribed at discharge, all ave- 859
among the groups. However, there was a practice
804 nues that lead to opioid use reduction must be 860
change at the research facility during enrollment,
805 explored. In addition, the lack of a significant dif- 861
and all women were prescribed NSAIDs on a
806 ference in opioid use on Days 3 or 4 among the 862
timed schedule, rather than on an as-needed
807 groups calls into question the sustained relief 863
basis. Consequently, we were unable to deter-
808 brought about by the use of this therapy and 864
mine what effect, if any, group assignment had on
809 highlights the need for additional research on the 865
the participants’ NSAID use.
810 866
811 867
812 868
813 869
814 870
815 871
816 872
817 873
818 874
819 875
820 876
821 877
822 878
823 879
824 880
825 881
826 882
827 883
828 884
829 885
830 886
831 887
832 888
833 889
834 890
Figure 3. Stress before (Pre) and after (Post) intervention. Change in stress was significant in response to the massage
835 891
(intercept [i] ¼ 3.69  0.23, t ¼ 16.10, p < .001 and slope [s] ¼ –2.30  0.25, t ¼ –9.39, p < .001) and individualized attention
836 892
interventions (i ¼ 2.15  0.27, t ¼ 7.89, p < .001 and s ¼ –0.28  0.12, t ¼ –2.46, p ¼ .014) but not to standard care (i ¼ 2.68 
837 893
0.28, t ¼ 9.54, p < .001 and s ¼ –0.02  0.18, t ¼ –0.10, p ¼ .918). Even when baseline differences were considered (t ¼ 4.4,
838 894
p < .001), improvement in stress over time differed significantly among the three arms, favoring the massage intervention (t ¼
839 895
7.2, p < .001).
840 896

8 JOGNN, -, -–-; 2018. https://doi.org/10.1016/j.jogn.2018.07.006 http://jognn.org

FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce

 Simonelli, M. C. et al. RESEARCH

897 953
898 954
899 955
900 956
901 957
902 958
903 959
904 960
905 961
906 962
907 963
908 964
909 965
910 966
911 967
912 968
913 969
914 970
915 971
916 972
917 973
Figure 4. Relaxation before (Pre) and after (Post) intervention. Change in relaxation was significant in response to the
918 974
massage intervention (intercept [i] ¼ 4.56  0.28, t ¼ 16.50, p < .001 and slope [s] ¼ –3.10  0.32, t ¼ –9.80, p < .001) but not
919 975
to standard care (i ¼ 3.33  0.31, t ¼ 10.66, p < .001 and s ¼ –0.46  0.32, t ¼ –1.45, p ¼ .146) or to the individualized
920 976
attention intervention (i ¼ 3.55  0.31, t ¼ 11.35, p < .001 and s ¼ –0.08  0.23, t ¼ –0.36, p ¼ .718). Even when baseline
921 977
differences were considered (t ¼ 2.9, p ¼ .003), improvement in relaxation over time differed significantly among the three arms
922 978
favoring the massage intervention (t ¼ 5.8, p < .001).
923 979
924 980
925 981
926 982
927 983
928 984
929 985
930 986
931 987
932 988
933 989
934 990
935 991
936 992
937 993
938 994
939 995
940 996
941 997
942 998
943 999
944 1000
945 1001
946 1002
947 1003
948 1004
Figure 5. Opioid use for each postoperative day of the hospitalization. Opioid use was less in the massage group compared
949 1005
with the other trial arms on Postoperative Day 1 (t ¼ 2.2, p ¼ .031), but there were no differences in the improvement of daily
950 1006
opioid use among the three groups over time (t ¼ 0.6, p ¼ .559). BMI ¼ body mass index; MG ¼ milligram; POD ¼ post-
951 1007
operative day.
952 1008

JOGNN 2018; Vol. -, Issue - 9

FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce

 RESEARCH Connective Tissue Massage After Cesarean Birth

1009 optimal dosage of massage needed to provide the VAS scores and opioid use. Incorporating the VAS 1065
1010 most favorable pain management regimen. in standard nursing assessment, rather than solic- 1066
1011 iting a verbal pain score from women, may improve 1067
1012 It is also possible that the reduction in opioid use the accuracy of pain assessments and the quality 1068
1013 seen in all groups from Day 1 to Day 4 is a of pain management strategies. 1069
1014 reflection of the fact that most participants were 1070
1015 discharged on Day 4, and the dosage docu- Clinical Implications 1071
1016 mented in the medical record may not have Although previous researchers investigated mas- 1072
1017 included the full 24-hour opioid consumption for sage as an intervention to decrease pain, including 1073
1018 participants. In future studies, researchers might postoperative pain, we needed to collect further 1074
1019 explore posthospitalization opioid use to expli- evidence with regard to the efficacy of massage to 1075
1020 cate further any sustained effect of massage. In reduce pain after unplanned cesarean birth. Given 1076
1021 addition, it is important to note that the partici- that all participants experienced the effects of labor 1077
1022 pants in the individualized attention group had and subsequent muscle strain and tension, we used 1078
1023 decreased pain and stress between Time 1 and head, neck, and back massage as our intervention. 1079
1024 Time 2 when compared with standard of care We also examined the intervention’s effect on opioid 1080
1025 participants. Perhaps giving women the oppor- administration and controlled for the potential ef- 1081
1026 tunity to share their birth stories with their nurses fects of individualized attention on postoperative 1082
1027 should be considered as an additional interven- pain management. The incorporation of these 1083
1028 tion. Future research to examine multimodal pain measures strengthened the quality of the research 1084
1029 management strategies is critical to improving and provided data on the effectiveness of head, 1085
1030 care for women and their newborns. neck, and back massage as a nonpharmacologic 1086
1031 pain management technique. The findings of re- 1087
1032 Limitations ductions of pain and stress in the individualized 1088
1033 Despite randomization, participants’ reports of attention group also suggest that nurses should take 1089
1034 stress on the VAS were significantly different at time to debrief birth experiences with women. 1090
1035 Time 1. Participants randomized to the massage 1091
1036 group reported more stress before the intervention 1092
Conclusion
1037 than the standard of care and the individualized 1093
The use of massage as an integrative post-
1038 attention groups. We informed participants of their 1094
operative pain intervention reduced pain and
1039 group assignment before they completed the VAS 1095
stress and increased relaxation for primiparous
1040 at Time 1. Perhaps in a future study, participants 1096
women after unplanned cesarean births. The
1041 should complete the initial survey before disclosure 1097
importance of alternative pain management stra-
1042 of group assignment, which would thus remove any 1098
tegies and decreased opioid administration
1043 potential bias associated with this knowledge. The 1099
cannot be overstated for this population. Post-
1044 addition of biomarkers for pain and stress would 1100
partum women need to remain alert and awake to
1045 also strengthen our ability to assess these experi- 1101
provide effective care for and feeding of their
1046 ences more objectively and develop potential 1102
newborns, develop positive maternal–infant
1047 mechanisms for stress and pain reduction. 1103
bonds, and return to optimal health. Our findings
1048 1104
expand the knowledge of the efficacy of comple-
1049 The demographics of the sample, although repre- 1105
mentary therapies, specifically massage, as an
1050 sentative of the population in the city where we 1106
effective pain management strategy for women
1051 conducted the research, were not diverse. Future 1107
who experience unplanned cesarean births.
1052 research with a more diverse population and the 1108
1053 incorporation of cultural influences on pain, stress, 1109
1054 and relaxation is important. We also need to study Acknowledgment 1110
1055 pain assessment strategies. The current practice is Supported by a grant from the Yvonne L. Munn Q12
1111
1056 to verbally solicit a number from 0 (no pain) to 10 Center of Nursing Research, Massachusetts 1112
1057 (worst pain imaginable). The verbal number General Hospital. The authors thank undergrad- 1113
1058 assignment from an individual can be arbitrary, uate research fellows Kimberlyn Austin, Katherine 1114
1059 culturally influenced, and neither correlated with Hohne, and Kristin Sullivan. 1115
1060 the individual’s objective signs of discomfort nor 1116
1061 with his/her request for medication. Indeed, the 1117
1062 average daily pain scores did not statistically differ REFERENCES 1118
1063 based on group assignment in our study, despite Abbaspoor, Z., Akbari, M., & Najar, S. (2014). Effect of foot and hand 1119
1064 the significant differences in the postintervention massage in post-cesarean section pain control: A randomized 1120

10 JOGNN, -, -–-; 2018. https://doi.org/10.1016/j.jogn.2018.07.006 http://jognn.org

FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce

 Simonelli, M. C. et al. RESEARCH

1121 control trial. Pain Management Nursing, 15, 132–136. https:// Jones, C. M. (2013). Heroin use and heroin use risk behaviors among 1174
1122 doi.org/10.1016/j.pmn.2012.07.008 nonmedical users of prescription opioid pain relievers—United 1175
Adams, R., White, B., & Beckett, C. (2010). The effects of massage for States, 2002–2004 and 2008–2010. Drug and Alcohol Dependence,
1123 1176
pain management in the acute care setting. International Jour- 132(1–2), 95–100. https://doi.org/10.1016/j.drugalcdep.2013.01.007
1124 nal of Therapeutic Massage & Bodywork, 3, 4–7. https://doi.org/ Karlstrom, A. (2007). Postoperative pain after cesarean birth affects
1177
1125 10.3822/ijtmb.v3i1.54 breastfeeding and infant care. Journal of Obstetric, Gyneco- 1178
1126 Bauer, B. A., Cutshall, S. M., Wentworth, L. J., Engen, D., Messner, P. logic, & Neonatal Nursing, 36, 430–440. https://doi.org/10.1111/ 1179
1127 K., Wood, C. M., … Sundt, T. M. (2010). Effect of massage j.1552-6909.2007.00160.x Q10 1180
therapy on pain, anxiety, and tension after cardiac surgery: A Kohl, A., Winfried, R., & Glombiewski, J. A. (2013). Acceptance,
1128 1181
randomized study. Complementary Therapies in Clinical Prac- cognitive restructuring, and distraction as coping strategies for
1129 tice, 16(2), 70–75. https://doi.org/10.1016/j.ctcp.2009.06.012 acute pain. Journal of Pain, 14, 305–315. https://doi.org/10.
1182
1130 Betrán, A. P., Ye, J., Moller, A.-B., Zhang, J., Gümezoglu, A. M., & 1016/j.jpain.2012.12.005 1183
1131 Torloni, M. R. (2016). The increasing trend in cesarean section Kolcaba, K. Y. (1992). Holistic comfort. Advances in Nursing Science, 1184
1132 rates: Global, regional, and national estimates: 1990–2014. 15(1), 1–10. https://doi.org/10.1097/00012272-199209000-00003 1185
PLOS ONE, 11(2), e0148343. https://doi.org/10.1371/journal. Kolcaba, K. Y. (1994). A theory of holistic comfort for nursing. Journal
1133 1186
pone.0148343 of Advanced Nursing, 19, 1178–1184. https://doi.org/10.1111/j.
1134 Bijur, P., Silver, W., & Gallagher, J. (2001). Reliability of the visual 1365-2648.1994.tb01202.x
1187
1135 analog scale for measurement of acute pain. Academic Emer- Malacca Securities. (n.d.) MPlus (version 8) [Software]. Los Angeles, 1188
1136 gency Medicine, 8(12), 1153–1157. https://doi.org/10.1111/j. CA: Malacca Securities. 1189
1137 1553-2712.2001tb01132.x Martin, J. A., Hamilton, B. E., Osterman, M. K., Driscoll, A. L., & Drake, 1190
Boyd, C., Crawford, C., Paat, C. F., Price, A., Xenakis, L., & Zhang, W. P. (2018). Births: Final data for 2016. National Vital Statistics
1138 1191
(2016). The impact of massage therapy on function in pain Reports, 67(1), 1–55.
1139 populations—A systematic review and meta-analysis of ran- McRee, L., Pasvogel, A., Hallum, A. V., Behr, S. E., Garcia, F. A. R., &
1192
1140 domized controlled trials: Part III, surgical pain populations. Loeb, R. G. (2007). Effects of preoperative massage on intra- 1193
1141 Pain Medicine, 17, 1757–1772. https://doi.org/10.1093/pm/ and postoperative outcomes. Journal of Gynecologic Surgery, 1194
1142 pnw100 23(3), 97–103. https://doi.org/10.1089/gyn.2007.B-02276-1 1195
Chuntharapat, S., Petpichetchian, W., & Hatthakit, V. (2008). Yoga Modarres, M., Afrasiabi, S., Rahnama, P., & Montazen, A. (2012).
1143 1196
during pregnancy: Effects on maternal comfort, labor pain, and Prevalence and risk factors of childbirth-related post-traumatic
1144 birth outcomes. Complementary Therapies in Clinical Practice, stress symptoms. BMC Pregnancy and Childbirth, 12(88), 1–6.
1197
1145 14, 105–115. https://doi.org/10.1016/j.ctcp.2007.12.007 https://doi.org/10.1186/1471-2393-12-88 1198
1146 Clark, E., & Silver, R. (2011). Long-term maternal morbidity associated Saatsaz, S., Rezaei, R., Alipour, A., & Beheshti, Z. (2016). Massage as 1199
1147 with repeat cesarean delivery. American Journal of Obstetrics adjuvant therapy in the management of post-cesarean pain and 1200
and Gynecology, 205(6), S2–S10. https://doi.org/10.1016/j. anxiety: A randomized clinical trial. Complementary Therapies
1148 1201
ajog.2011.09.028 in Clinical Practice, 24(8), 92–98. https://doi.org/10.1016/j.ctcp.
1149 Cutshall, S., Wentworth, L., Engen, D., Sundt, T., Kelly, R., & Bauer, B. 2016.05.014
1202
1150 (2010). Effect of massage therapy on pain, anxiety, and tension Sakalis, V., William, T., Hepworth, G., Hartmann, P., & Tamimi, G. 1203
1151 in cardiac surgical patients: A pilot study source. Comple- (2013). A comparison of early sucking dynamics during breast 1204
1152 mentary Therapies in Clinical Practice, 16(2), 92–95. https://doi. feeding after cesarean section and vaginal birth. Breastfeeding 1205
org/10.1016/j.ctcp.2009.10.006 Medicine, 8, 79–85. https://doi.org/10.1089/bfm.2012.0018
1153 1206
Dehcheshmeh, F. S., & Rafiei, H. (2015). Complementary and alter- Sprenger, C., Eippert, F., Finsterbusch, J., Bingel, U., Rose, M., &
1154 native therapies to relieve labor pain: A comparative study Buchel, C. (2012). Attention modulates spinal cord responses
1207
1155 between music therapy and Hoku point ice massage. Com- to pain. Current Biology, 22, 1019–1022. https://doi.org/10. 1208
1156 plementary Therapies in Clinical Practice, 21, 229–232. https:// 1016/j.cub.2012.04.006 1209
1157 doi.org/10.1016/j.ctcp.2015.09.002 Taghinejad, H., Delpisheh, A., & Suhrabi, Z. (2010). Comparison be- 1210
 lu, T. (2017). The effect of nursing care based on
Derya, V., & Pasinliog tween massage and music therapy to relive the severity of labor
1158 1211
comfort theory on women’s postpartum comfort levels after pain. Women’s Health, 6, 377–381. https://doi.org/10.2217/
1159 cesarean section. International Journal of Nursing Knowledge, WHE.10.15
1212
1160 28(3), 138–144. https://doi.org/10.1111/2047-3095.12122 Turchaninov, R. (2011). How massage therapy heals the body. Part III: 1213
1161 Dion, L., Rodgers, N., Cutshall, S. M., Cordes, M. E., Bauer, B., Cas- Vasodilation mechanisms. Journal of Massage Science, 3(3), 4–6. 1214
1162 sini, S. D., & Cha, S. (2011). Effect of massage on pain man- Wentworth, L. J., Brieses, L. J., Timini, F. K., Sanvick, C. L., Bartel, D. 1215
agement for thoracic surgery patients. International Journal of C., Cutshall, S. M., … Bauer, B. A. (2009). Massage therapy
1163 1216
Therapeutic Massage and Body Work, 4(2), 2–6. https://doi.org/ reduces tension, anxiety, and pain in patients awaiting invasive
1164 10.3822/ijtmb.v4i2.100 cardiovascular procedures. Progress in Cardiovascular
1217
1165 Express Scripts. (2014). A nation in pain: Focusing on U.S. opioid Nursing, 24, 155–161. https://doi.org/10.1111/j.1751-7117. 1218
1166 trends for treatment of short-term and longer-term pain. 2009.00054.x 1219
1167 Retrieved from http://lab.express-scripts.com/lab/publications/ Xue, M., Fan, L., Ge, L. N., Zhang, Y., Ge, J. L., Gu, J., … Chen, Y. 1220
a-nation-in-pain. (2016). Postoperative foot massage for patients after cesarean
1168 1221
Hanan, A., Kamilla, R., Ahmed, R., & Amina, M. (2014). Investigate the delivery. Zeitschrift für Geburtshilfe und Neonatologie, 220(4),
1169 utilization of natural measures on relieving post cesarean inci- 173–178. https://doi.org/10.1055/s-0042-104802
1222
1170 sion pain. Asian Journal of Nursing Education and Research, 4, Zanardo, V., Sveglido, G., Cavallin, F., Giustardi, A., Cosmi, E., Litta, P., 1223
1171 388–393. & Trevisanuto, D. (2010). Elective cesarean delivery: Does it 1224
1172 Ingrande, J., & Lemmens, H. (2010). Dose adjustment of anesthetics have a negative effect on breastfeeding? Birth Issues in Peri- 1225
in the morbidly obese. British Journal of Anaesthesia, 105(S1), natal Care, 37(4), 275–279. https://doi.org/10.1111/j.1523-
1173 1226
16–23. https://doi.org/10.1093/bja.aeq312 536X.2010.00421.x

JOGNN 2018; Vol. -, Issue - 11

FLA 5.5.0 DTD
JOGN375_proof
10 August 2018
12:33 pm
ce