Does muscle-sparing thoracotomy as opposed to posterolateral thoracotomy result in better recovery?

Abstract

A best evidence topic was written according to a structured protocol. The question addressed was whether muscle-sparing thoracotomy (MST), as opposed to posterolateral thoracotomy (PLT), results in better recovery. A total of 108 papers were found using the reported searches of which eight represented the best evidence to answer the clinical question. The authors, date, journal, study type, population, main outcome measures and results are tabulated. A recent large prospective, randomized, double-blinded, controlled study demonstrated a shorter length of stay in patients undergoing MST. It failed to demonstrate any significant difference in pain reported or pulmonary function. A separate prospective randomized controlled trial focussed on pain, pulmonary function, late shoulder range of motion and late muscle strength. It failed to show any significant difference in these domains between PLT and MST. While the mean ‘opening time’ is greater when performing a MST, this is negated by a shorter mean ‘closing time’ when compared with PLT. Overall, the evidence suggests that MST results in greater early (1 week) preservation of skeletal muscle strength and range of motion over PLT. This difference has disappeared at 1 month. There is little evidence to suggest a difference in pulmonary function or pain dependent on the thoracotomy type. Moreover, analgesic consumption is similar. However, there is an inverse relationship between the incision length and the post-thoracotomy syndrome.

INTRODUCTION

A best evidence topic was constructed according to a structured protocol. This protocol is fully described in the ICVTS [1].

CLINICAL SCENARIO

When starting a new job as a registrar in thoracic surgery, you notice that your consultant takes care to cause minimal damage to the serratus anterior and lattisimus dorsi while performing a thoracotomy. As this differs from what you have previously been taught, you enquire as to his reasoning. He asserts that by sparing the muscles he preserves strength, enhances recovery and reduces pain. You resolve to review the literature.

THREE-PART QUESTION

In [patients undergoing thoracotomy] does a [muscle-sparing approach] result in [preserved strength/enhanced recovery/reduced pain]?

SEARCH STRATEGY

Search strategy using Medline from 1950 to May 2011 with the PubMed interface (muscle-sparing thoracotomy OR auscultatory triangle thoracotomy OR minimally invasive thoracotomy) AND (posterolateral thoracotomy OR conventional thoracotomy) AND (pain or strength or recovery).

SEARCH OUTCOME

One hundred and eight papers were found using the reported search on PubMed. From these, eight papers were identified that provided the best evidence to answer the question. These are presented in Table 1.

Table 1:

Best evidence papers

Author, date, journal and country Study type (level of evidence)	Patient group	Outcomes	Key results	Comments
Nosotti et al. (2010), Interact CardioVasc Thorac Surg, Italy [2] Prospective, randomized, controlled and double-blinded study (level 2)	All patients scheduled for thoracotomy and pulmonary lobectomy for stage I or II non-small-cell lung cancer between July 2003 and 2006 within a single centre were included and were followed up for 3 years Of 378 consecutive patients undergoing pulmonary lobectomy, 107 consecutive patients were eligible for the study Fifty patients were randomly assigned to the MST group and 50 to the PLT group All operations were carried out by the same surgeon The two groups were similar in age, gender, body mass index, preoperative functional class and stage Both the investigators and patients were blinded to the study treatment until the end of follow-up	Operative time (min) Intercostal incision length (cm) Ribs spreading (cm) Rib fracture Tumour diameter (cm) No. of lymphatic stations dissected Postoperative stay Postoperative complications (total) Aggregate analgesic score (AAS)	MST group (n = 50) vs PLT group (n = 50) 176 ± 27 vs 180 ± 38 (P = 0.6) 15.1 ± 2.7 vs 16.1 ± 2.6 (P = 0.08) 7.3 ± 1.4 vs 7.4 ± 1.2 (P = 0.7) 7 vs 5 (P = 0.7) 3.1 ± 1.7 vs 2.9 ± 1.2 (P = 0.5) 5.6 ± 1 vs 5.9 ± 1.2 (P = 0.2) 6.6 ± 2.1 vs 8.2 ± 3.1 (P = 0.003) 7 vs 11 (P = 0.4) 159 ± 18 vs 173 ± 15 (P = 0.0001)	This prospective, randomized, double blinded prospective study demonstrated MST was associated with significantly less analgesic requirement and a shorter postoperative stay Differences in the mean daily values of the visual analogue scale, used to assess pain severity, did not reach statistical significance Incision length and rib spreading width had an inverse correlation with the postoperative pain on the first postoperative day (P = 0.0072 and P = 0.0355) Rib fractures did not affect the perception of early postoperative pain Incision length had an inverse correlation with AAS (P = 0.0058) An inverse correlation between incision length and post-thoracotomy syndrome was observed (P = 0.0019) Matched-pairs analysis demonstrated a significant worse result of the adduction strength in the PLT group during the hospital stay (P = 0.0004) Early and late postoperative pulmonary function tests did not disclose any significant disparity
Athanassiadi et al. (2007), Eur J Cardiothorac Surg, Greece [3] Prospective randomized study (level 2)	Hundred patients operated on from June through December 2004 Fifty patients had MST of 6–8 cm (performed by consultant A) and 50 had a PLT of more than 8 cm with division of latissimus dorsi and serratus anterior muscles (performed by consultant B) Operations performed were atypical resections and lobectomies There was no difference in demographics, tumour stage and type of lung resection Postoperative pain (quantitated by the visual analogue scale), preoperative and postoperative pulmonary function, shoulder strength and range of motion were evaluated	Mean daily pain values–visual analogue scale POD 2 POD 8 POD 30 POD 60 Pulmonary function tests Vital capacity (l) Preoperative 1 week 1 month 2 months FEV1 (l) Preoperative 1 week 1 month 2 months Range of motion Internal rotation (0–90°) Preoperative 1 week 1 month 2 months Intra- and postoperative data Operating time (min) Blood loss (ml) Chest tube drainage (ml)	MST (n = 50) vs PLT (n = 50) 5.8 ± 0.9 vs 5.1 ± 0.6 (P < 0.05) 2.4 ± 1.2 vs 2.2 ± 1.0 (N.S.) 1.5 ± 1.1 vs 1.4 ± 0.9 (N.S.) 1.2 ± 1.0 vs 1.1 ± 1.0 (N.S.) 3.15 ± 0.93 vs 2.97 ± 0.81 (N.S.) 2.61 ± 0.48 vs 2.46 ± 0.62 (N.S.) 2.77 ± 0.68 vs 2.59 ± 0.57 (N.S.) 2.91 ± 0.74 vs 2.73 ± 0.86 (N.S.) 2.61 ± 0.76 vs 2.38 ± 0.83 (N.S.) 2.05 ± 0.68 vs 1.81 ± 0.78 (N.S.) 2.16 ± 0.53 vs 1.89 ± 0.67 (N.S.) 2.27 ± 0.58 vs 1.98 ± 0.70 (N.S.) 83 ± 11.7 vs 83 ± 10.8 (N.S.) 45 ± 7.0 vs 59 ± 6.7 (P < 0.05) 77 ± 6.4 vs 76 ± 7.2 (N.S.) 82 ± 9.4 vs 82 ± 10.4 (N.S.) 251 ± 86 vs 263 ± 67 514 ± 233 vs 387 ± 284 647 ± 312 vs 536 ± 257	Pain reported during hospitalization and after hospital discharge within 8, 30 and 60 days did not differ within the two groups Pulmonary function was not influenced by the type of incision Analysis of shoulder motion range demonstrated a slight but statistically significant difference between groups and favouring the MST only within the first week Flexion, abduction and external and internal rotation returned to preoperative values by the first month No blood transfusions were required and operating time was similar between groups The authors commented that an advantage of MST over PLT is the preservation of chest wall musculature. This permits future rotational muscle flaps along with a better cosmetic result A limitation of the study is that the groups are operated on by two different surgeons. It is impossible to ascertain to what degree inter-group similarities and differences are due to a variation in individual abilities and performance A limitation in this paper is the lack of clarity regarding who performed the operation. It is not possible to determine if one or multiple surgeons operated and to what extent this impacted on the reported outcomes
Ziyade et al. (2010), Thorac Cardiovasc Surg, Turkey [4] Prospective randomized study (level 2)	From January 2003 to December 2004, 50 patients with a diagnosis of primary lung cancer underwent pulmonary resection were randomized to PLT vs MST There were no significant differences in terms of demographic data, body mass index, isokinetic and respiratory muscle strength, maximal inspiratory pressure (MIP) and maximal expiratory pressure (MEP) between the two groups preoperatively The groups were compared in terms of shoulder abduction/adduction isokinetic muscle strength and respiratory muscle strength	Peak torque 60°/s abduction Peak torque 120°/s abduction Peak torque 180°/s abduction Total work 60°/s abduction Total work 120°/s abduction Total work 180°/s abduction Maximal inspiratory pressure (MIP) Maximal expiratory pressure (MEP)	Mean values of decrease PLT (n = 25) vs MST (n = 25) 6.79 ± 8.14 vs 5.63 ± 9.20 (P = 0.66) 9.80 ± 12.77 vs 8.23 ± 13.05 (P = 0.689) 10.40 ± 16.21 vs 10.68 ± 19.29 (P = 0.958) 40.50 ± 42.57 vs 49.60 ± 63.51 (P = 0.577) 32.54 ± 42.16 vs 20.63 ± 49.79 (P = 0.396) 20.56 ± 39.32 vs 28.95 ± 52.48 (P = 0.549) 8.74 ± 13.55 vs 4.19 ± 15.87 (P = 0.038) 12.62 ± 15.76 vs 7.61 ± 27.93 (P = 0.013)	The MST group achieved its preoperative values of peak torque earlier than the PLT group MST had less effect on total work compared with PLT The decrease in postoperative (at 3 months) MIP and MEP was statistically significant in those undergoing PLT but not those with MST The differences in preoperative and postoperative (at 3 months) MEP and MIP were significantly higher in the PLT group The authors concluded that for the preservation of muscle strength, especially in patients whose jobs involved manual work, MST should be the first choice rather than PLT. Moreover, if necessary, the latissimus dorsi muscle can be used more extensively as a flap after MST A limitation in this paper is the lack of clarity regarding who performed the operation. It is not possible to determine if one or multiple surgeons operated and to what extent this impacted on the reported outcomes
Akçali et al. (2003), Ann Thorac Surg, Turkey [5] Prospective, randomized, blinded study (level 2)	Sixty patients requiring thoracotomy for a variety of thoracic diseases (neoplastic and non-neoplastic) were randomly selected to undergo either PLT (n = 30 patients) or a MST (n = 30) between 1999 and 2000 There were no significant differences between the two groups with respect to age, gender, diagnosis or the procedure performed The 2 groups were compared in terms of surgical approach time, postoperative pain (quantitated by narcotic requirements and the visual analogue scale), pulmonary function, shoulder strength and range of motion On 10 patients from each group, a needle electroneuromyography was performed at 1 month postoperatively using a bipolar recording electrode positioned on the latissimus dorsi and serratus anterior muscle groups	Pulmonary function tests VC Preoperative 1 week 1 month Muscle strength (0–5) Latissimus dorsi Preoperative 1 week 1 month Serratus anterior Preoperative 1 week 1 month Surgical approach time (min) Opening time Closing time Real operating time Visual analogue scale Day 1 Day 2 Day 4 Day 8 Narcotic requirement (mg/day) Day 1 Day 2 Day 4 Day 8	PLT (n = 30) vs MST (n = 30) 2.9 ± 1.1 vs 2.7 ± 0.7 (P = 0.40) 2.3 ± 0.9 vs 2.4 ± 0.6 (P = 0.6145) 2.5 ± 0.9 vs 2.5 ± 0.6 (P = 1.00) 4.7 ± 0.4 vs 4.8 ± 0.4 (P = 0.34) 3.2 ± 0.8 vs 4.6 ± 0.6 (P < 0.001) 4.5 ± 0.7 vs 4.7 ± 0.5 (P = 0.21) 4.7 ± 0.4 vs 4.7 ± 0.5 (P = 1.00) 3.0 ± 1.0 vs 4.5 ± 0.6 (P < 0.001) 4.5 ± 0.7 vs 4.6 ± 0.6 (P = 1.00) 33.9 ± 6.58 vs 42.03 ± 5.59 (P < 0.0001) 49.9 ± 3.83 vs 30.23 ± 5.78 (P < 0.0001) 83.8 ± 7.21 vs 72.26 ± 10.64 (P < 0.0001) 4.33 ± 0.75 vs 3.90 ± 0.88 (P = 0.046) 4.16 ± 0.79 vs 3.66 ± 0.80 (P = 0.02) 3.66 ± 0.92 vs 3.06 ± 1.01 (P = 0.02) 3.2 ± 1.06 vs 2.03 ± 0.99 (P < 0.0001) 278.3 ± 26.53 vs 243.3 ± 41.36 (P = 0.0003) 232.3 ± 40.52 vs 200.3 ± 45.27 (P = 0.0055) 211.6 ± 59.0 vs 165.3 ± 53.48 (P = 0.0023) 182.8 – 55.0 vs 117.6 – 48.34 (P < 0.0001)	The range of shoulder motion returned to preoperative values of range by 2 weeks in the MST group, they returned to normal by 1 month in the PLT group In MST group, shoulder strength appeared to have been preserved in the latissimus dorsi and serratus anterior muscles at the 1-week postoperative assessment Shoulder girdle strength returned to preoperative levels of strength by 1 month in both groups Although it took an average of 9 min longer to open the chest with MST, this time was regained through a quicker closure that did not require suture reapproximation of the thoracic musculature Electromyographic studies revealed that 2 patients (20%, n = 10) in the PLT group had moderate neurogenic damage, and 8 had severe damage (80%); 1 of them had winged scapula. However, 2 patients (20%, n = 10) in the MST group had mild neurogenic damage, and 2 had moderate damage (20%) A limitation in this paper is the lack of clarity regarding who performed the operation. It is not possible to determine if one or multiple surgeons operated and to what extent this impacted on the reported outcomes
Sugi et al. (1996), World J Surg, Japan [6] Prospective randomized study (level 2)	Thirty patients with primary lung cancer (stage I or II) were randomized to PLT (PLT group, n = 15) or MST (MST group, n = 15) All patients underwent resection of the primary lobe and mediastinal lymphadenectomy in the same manner by the same surgeon There were no significant differences between groups with respect to age, sex or lung cancer stage	Operating time (min) Opening time (min) Closing time (min) Operative field size (cm2) Number of dissected lymph nodes Pulmonary function % FVC Before surgery After surgery % FEV1 Before surgery After surgery	PLT Group (n = 15) vs MST Group (n = 15) 66 ± 12 vs 87 ± 13 (P < 0.05) 38 ± 12 vs 66 ± 12 (P < 0.05) 40 ± 8 vs 38 ± 5 (P > 0.05) 218 ± 31 vs 165 ± 41 (P < 0.05) 22 ± 1 vs 23 ± 2 (P > 0.05) 82 ± 9 vs 85 ± 6 (P > 0.05) 66 ± 9 vs 63 ± 6 (P > 0.05) 92 ± 18 vs 95 ± 9 (P > 0.05) 68 ± 18 vs 71 ± 9 (P > 0.05)	Mean daily estimates of pain using the visual analogue scale (VAS) in the MST group were less than those in the PLT at all time points. These differences were statistically significant at 1, 3 and 5 days after surgery. The average requirement for narcotics in the MST group was significantly less than that in the PLT group at 3 and 5 days after surgery The shoulder range of flexion in the sagittal plane was significantly reduced at 14 days after surgery in the PLT group. The range-of-motion in the MST group was significantly better than that in the PLT group. In other planes, the range of motion was significantly reduced in both groups for 2 weeks after surgery but returned to baseline by 1 month after surgery A postoperative seroma occurred in one patient in the MST group
Benedetti et al. (1998), J Thorac Cardiovasc Surg, Italy [7] Prospective cohort study (level 3)	The study was performed in 24 patients; PLT was performed in 11 patients (4 had primary lung cancer and 7 had metastatic disease); MST was performed in 13 patients (7 had primary lung cancer and 6 had metastatic disease) Neurophysiological recordings were performed 1 month after surgery to assess the presence of the superficial abdominal reflexes (mediated in part by the intercostal nerves), the somatosensory-evoked responses after electrical stimulation of the surgical scar and the electrical thresholds for tactile and pain sensations of the surgical incision The presence of post-thoracotomy pain was assessed by means of a numerical rating scale ranging from 0 = no pain to 10 = unbearable pain and a VAS	Mean reduction in amplitude of superficial abdominal reflexes Mean reduction in amplitude of somatosensory-evoked potentials Percentage increase in tactile threshold Percentage increase in pain threshold Mean VAS	PLT (n = 11) vs MST (n = 13) 64.6% ± 49.4% SD vs 5.3% ± 31.9% SD (P = 0.01) 37.8% ± 15.4% SD vs 14.6 ± 16.3% SD (P = 0.004) 357.7% ± 130.2% SD vs 194.3% ± 109.5% SD (P = 0.005) 172.5% ± 29.4% SD vs 109.8% ± 37.2% SD (P = 0.0005) 21 ± 17.67 SD vs 5.85 ± 8.11 SD (P = 0.01)	Recording of the superficial abdominal reflexes 1 month after operation showed that in 7 of 11 PLT patients (63.6%) and in only 1 MST patient (7.7%) the ipsilateral electromyography responses were completely absent The skin of the operative scar was hypaesthesic and hypalgesic in both groups 1 month after operation. This was less severe in MST compared with PLT patients, indicating less severe nerve damage after MST A limitation in this paper is the lack of clarity regarding who performed the operation. It is not possible to determine if one or multiple surgeons operated and to what extent this impacted on the reported outcomes Furthermore, there is no suggestion of randomization to groups
Khan et al. (2000), Eur J Cardiothorac Surg, N. Ireland [8] Matched case control study (level 3)	Ten patients who had undergone MST at least 1 year previously were matched with 10 patients who had undergone PLT over the same period Each pair was matched for age, sex, dominant hand, side of the operation, time since operation and presence or absence of history of previous muscle training Latissimus dorsi muscle strength was assessed by testing the shoulder adduction strength through an arc of 90 ± 08 using isokinetic technique Early postoperative pain was assessed indirectly by calculating the analgesic requirement in the first 5 postoperative days (CAS) An assessment of chronic post-thoracotomy pain was made using a questionnaire presented to the patients at the time of muscle testing (WSS)	Mean CAS Mean WSS Shoulder abduction Strength ratio (operated/non-operated)	PLT (n = 10) vs MST (n = 10) (Mean value) 190 vs 221 (P = 0.60) 7.2 vs 6.4 (P = 0.62) 78% vs 102% (P = 0.04)	All patients reported at least one chronic post-thoracotomy symptom There was no significant difference between the two groups in terms of acute or chronic wound pain and other long-term wound-related symptoms Shoulder adduction strength was 24% greater in MST than PLT A limitation in this paper is the lack of clarity regarding who performed the operation. It is not possible to determine if one or multiple surgeons operated and to what extent this impacted on the reported outcomes
Ochroch et al. (2005), Chest, USA [9] Retrospective cohort study (level 3)	Analysis of a database originally collected to determine the effect of the timing of epidural analgesia on long-term outcome after thoracotomy Patients presenting for lobectomy, segmentectomy or bilobectomy Eighty-two subjects underwent MST and 38 subjects underwent PLT during the 16-month accrual period	Average (±SD) pain score over last 24 h (postoperative day 1–5) Mean worst pain (postoperative day 1–5) Mean least pain (postoperative day 1–5)	PLT group (n = 38) vs MST (n = 82) 2.9 ± 0.47 vs 3.0 ± 0.49 5.3 ± 0.51 vs 5.5 ± 0.55 0.8 ± 0.28 vs 1.3 ± 0.22	Pain reported during hospitalization and after hospital discharge did not differ with respect to incision type There were no differences between the groups in the volume of epidural medication received, the number of patient-controlled boluses delivered or the number of patient demands made during the lockout phase of the patient-controlled epidural analgesia. There was also no difference between the groups in the amount of supplemental ketorolac and non-epidural opioids administered. The median length of hospital stay was 5 days for both groups Postoperative physical activity levels were significantly less than those reported preoperatively, with a trend toward better functioning in the MT groups after 8 weeks A limitation in this paper is the lack of clarity regarding who performed the operation. It is not possible to determine if one or multiple surgeons operated and to what extent this impacted on the reported outcomes

RESULTS

Nosotti et al. [2] conducted a prospective, randomized, controlled, double blind trial to compare muscle-sparing thoracotomy (MST) and posterolateral thoracotomy (PLT). Pain, analgesic consumption and post-thoracotomy pain syndrome were the primary endpoints, whereas morbidity plus shoulder and pulmonary functions were the secondary endpoints. There were no differences in postoperative pain results, although the PLT group consumed more analgesics. The type of incision had no effect on 3-year post-thoracotomy syndrome. The MST group had better immediate shoulder strength but postoperative pulmonary function and complications were similar.

Athanassiadi et al. [3] prospectively randomized 100 patients to compare the two incisions for postoperative pain and shoulder function during and after hospitalization. The two groups showed no differences in pain reported during hospitalization and after hospital discharge within 30 and 60 days. Early shoulder function (within a week) was shown to be better preserved in cases of MST. This difference did not persist at 1 month.

Ziyade et al. [4] randomized 50 patients with a diagnosis of primary lung cancer from January 2003 to December 2004. The aim was to compare the effects of PLT and MST on pulmonary and muscle strength. The two groups were compared in terms of shoulder abduction/adduction isokinetic muscle strength and respiratory muscle strength. Statistically significant differences were found when maximal expiratory pressure and maximal inspiratory pressure were compared preoperatively and postoperatively (at 3 months).

Akçali et al. [5] conducted a prospectively, randomized, blinded study involving 60 consecutive patients to compare PLT and MST techniques in terms of surgical approach time, postoperative pain (quantitated by narcotic requirements and the visual analogue scale), pulmonary function, shoulder strength and range of motion. There were no differences in operative surgical time, pulmonary function, shoulder range of motion, mortality or hospitalization time. Postoperative pain in the MST group was significantly less throughout the first postoperative week. Preoperative levels of muscle strength had returned by 1 month in both groups. The two groups were the same in terms of morbidity, with the exception of postoperative seromas, which showed a prevalence of 16.6% in the muscle-sparing group.

Sugi et al. [6] concluded from their prospective randomized study that the MST approach generated a smaller operative field (218 ± 31 vs 165 ± 41 cm²) and required more operative time (87 ± 13 min) than the PLT (66 ± 12 min). Differences in the number of dissected mediastinal lymph nodes were insignificant. During the early postoperative days, pain and restriction of shoulder flexion were significantly less in the MST group than in the PLT group. Pulmonary function was reduced in both groups at 1 month after surgery but without any significant differences between both groups.

Benedetti et al. [7] performed neurophysiological recordings 1 month after either PLT or MST and correlated the degree of intercostal nerve damage to the severity of long-lasting post-thoracotomy pain. The disappearance of the abdominal reflexes, a larger reduction in amplitude of the somatosensory-evoked potentials and a larger increase of the sensory thresholds to electrical stimulation for both tactile perception and pain, in patients who underwent a PLT, confirmed that PLT is associated with a higher degree of intercostals nerve impairment than MST. In addition, there was a high correlation between these neurophysiological parameters and the post-thoracotomy pain experienced by the patients 1 month after surgery, indicating a causal role for nerve impairment in the long-lasting post-thoracotomy pain.

Khan et al. [8] compared both incisions in terms of the difference between latissimus dorsi muscle strength, postoperative pain and chronic wound-related symptoms. At least one chronic post-thoracotomy symptom was reported by all patients. Differences in acute or chronic wound pain and other long-term wound-related symptoms were insignificant. There was 24% greater shoulder adduction strength in MST than PLT groups.

Ochroch et al. [9] performed a post-hoc analysis on a database originally generated to determine the effect of the timing of epidural analgesia on long-term outcome after thoracotomy. They sought to determine whether there was a difference in postoperative pain and recovery following PLT vs MST incisions. There were no differences between both incisions in the pain reported during hospitalization and after hospital discharge. Postoperative physical activity levels were significantly reduced compared with those reported preoperatively, with the MST groups functioning better after 8 weeks. The type of incision did not predict complications, morbidity or mortality.

CLINICAL BOTTOM LINE

While the mean ‘opening time’ is greater when performing a MST, this is negated by a shorter mean ‘closing time’ when compared with PLT. MST results in greater early (1 week) preservation of skeletal muscle strength and range of motion over PLT. This difference has disappeared at 1 month. There is little evidence to suggest a difference in pulmonary function or pain dependent on the thoracotomy type. Moreover, analgesic consumption is similar. However, there is an inverse relationship between incision length and post-thoracotomy syndrome.

Conflict of interest: none declared.