ORIGINAL ARTICLE

Risk Factors for Chronic Saphenous Neuralgia Following Coronary Artery Bypass Graft Surgery Utilizing Saphenous Vein Grafts Sameh M. Hakim, MD*; Samer N. Narouze, MD, PhD† *Department of Anesthesiology, Faculty of Medicine, Ain Shams University, Cairo, Egypt; Center for Pain Medicine, Summa Western Reserve Hospital, Cuyahoga Falls, Ohio, U.S.A.



& Abstract Objectives: The aim of this trial was to determine risk factors for chronic saphenous neuralgia (SN) following harvesting of the great saphenous vein (GSV) for coronary artery bypass graft (CABG) surgery. Methods: In a prospective observational trial, 526 patients with no history of chronic painful disorders or surgery in the lower limbs were followed up for 13 weeks after undergoing CABG surgery in which GSV grafts were used. The primary outcome measure was persistence of clinically significant pain of neuropathic type in the territory supplied by the saphenous nerve beyond 12 weeks after surgery. Results: Eighty-one (15.4%) patients consistently had probable neuropathic pain of clinically significant severity throughout the follow-up period and were labeled as suffering from chronic SN. Multivariable binary logistic regression analysis showed that younger age (OR, 0.92; 95% CI, 0.88–0.95; P-value, < 0.0001), female gender (OR, 2.28; 95% CI, 1.21–4.29; P-value, 0.011), higher body mass index (OR, 1.25; 95% CI, 1.17–1.35; P-value, < 0.0001), diabetes mellitus (OR, 2.13; 95% CI, 1.13–4.01; P-value, 0.020), distal-to-proximal dissection of the GSV (OR, 7.28; 95% CI, 3.62–14.66; P-value, < 0.0001), and closure of the leg wound

Address correspondence and reprint requests to: Sameh M. Hakim, MD, 15 Gamal Nooh Street, Heliopolis, Almaza, Postal Code 11341, Cairo, Egypt. E-mail: [email protected]. Submitted: January 21, 2014; Revised June 19, 2014; Revision accepted: July 23, 2014 DOI. 10.1111/papr.12246

© 2014 World Institute of Pain, 1530-7085/14/$15.00 Pain Practice, Volume , Issue , 2014 –

in two layers (OR, 3.28; 95% CI, 1.81–5.95; P-value, 0.0001) were independent risk factors for chronic SN. Conclusions: Chronic SN after CABG surgery utilizing GSV grafts is not uncommon. Risk factors identified in this trial are younger age, female gender, higher body mass index, diabetes mellitus, distal-to-proximal dissection of the GSV, and closure of the leg wound in two layers. & Key Words: nerve pain, neuralgia, saphenous neuralgia, coronary bypass surgery, complications, postoperative neuralgia, risk factors, neuropathic pain

INTRODUCTION In recent years, the annual rate of coronary artery bypass graft (CABG) surgery has been undergoing steady decline, while the rate of percutaneous coronary interventions has remained substantially unchanged. Nonetheless, CABG surgery still comprises a considerable caseload for the health care system with an estimated annual rate of 1,081 (95% CI, 1032–1133) CABG procedures per million adults in the United States.1 Originally performed with grafts obtained from the great saphenous vein (GSV), the use of arterial conduits is currently the preferred technique for CABG surgery owing to the longer survival of the graft.2 Grafting all target coronary vessels using arterial conduits, however, is not always feasible and the GSV remains an indispensable means for bypassing diseased coronary arteries, especially for vessels other than the left anterior descending branch.3

2 

HAKIM AND NAROUZE

Among the local complications that may ensue following graft harvesting from the GSV, saphenous neuralgia (SN) is perhaps the most distressing to the patient.4–8 This complication arises as a sequela to injury of the saphenous nerve, which assumes an intimate relation with the GSV, especially in the lower portion of the leg where it is most vulnerable to trauma.9,10 There is currently no standard definition for the term “saphenous neuralgia”. However, the condition has been variably described as a constellation of symptoms experienced in the cutaneous territory supplied by the saphenous nerve, usually following surgical trauma to the nerve or its branches. These symptoms include various combinations of hypoesthesia, anesthesia, paresthesia, dysesthesia, hyperesthesia, and pain.4–8,11 SN developing after GABG surgery tends to improve with time;4 however, the condition may persist for a considerable length of time in a proportion of patients and could have significant impact on their quality of life, especially if pain is a prominent feature.12 Owing to the lack of a modular definition for what should constitute SN, the actual incidence of this complication is difficult to estimate. On the other hand, the near absence of pain in a considerable proportion of patients indiscriminately labeled as suffering from SN4,5 seems rather contradictory. Therefore, it may be more appropriate to preserve the term for that form of neuropathy that occurs subsequent to trauma to the saphenous nerve in which pain associated with other somastosensory signs is a prominent feature. In this regard, there is evidence that approximately 15% of patients may experience moderate or severe pain in the distribution of the saphenous nerve that persists beyond 3 months following GABG procedures in which the GSV is harvested.12 The determinants of SN following harvesting of the GSV for CABG surgery are not clearly understood. Nevertheless, both patient and surgical factors seem to contribute to the occurrence of this untoward outcome.4,13,14 The aim of this trial was to determine risk factors for chronic SN following harvesting of the GSV for CABG surgery. For definition of chronic SN, we have adopted the criteria proposed by the International Association for the Study of Pain (IASP) Special Interest Group on Neuropathic Pain (NeuPSIG)15 to grade the level of certainty for identifying the pain as neuropathic. This required that the pain had a distinct topographic distribution compatible with known anatomy of the saphenous nerve, had a plausible temporal relation to

the surgical trauma, and had concordant negative or positive somatosensory signs on neurological examination. These criteria provided sufficient level of certainty to identify the pain as “probable neuropathic pain”. Owing to the lack of a gold-standard test for the condition, it was deemed unfeasible to subject the patients to further confirmatory testing to identify the pain as “definite neuropathic pain”. According to the NeuPSIG criteria,15 the levels “definite” and “probable” indicate that the diagnosis of neuropathic pain has been established. To identify the pain as chronic and to be clinically relevant, we required that it persisted for more than 12 weeks and that it was of sufficient severity so as to be regarded by patients as affecting their daily activity.

METHODS This prospective, observational study was conducted at Ain Shams University Hospitals, Cairo, Egypt, during the period from November 2010 to March 2013. The study was open-ended with the aim to end patient recruitment once the needed number of patients was followed up for the intended period. After approval of the institutional review board, all patients scheduled for CABG surgery, where GSV graft was to be used, were invited to participate in the study if they had no history of chronic painful conditions or surgery in the lower limbs. Patients with a history of peripheral neuropathy were excluded, as were those who had history of stroke or peripheral vascular disease involving the lower limbs. At the preoperative visit, a written informed consent was obtained from participants who were specifically questioned on any pain or abnormal sensations in the lower limbs. Patients were also examined objectively for any sensory abnormality in the territory supplied by the saphenous nerve. Technique of Harvesting the GSV Grafts The surgeon in charge of harvesting the venous grafts determined the extent of the incision according to the anticipated length of the grafts required. A continuous incision was used to access the GSV. The decision to proceed with dissection of the vein in a proximal-todistal direction or the reverse was left to the surgeon’s discretion. Dissection was carried out such that any obvious injury to the saphenous nerve or to any of its branches was to be avoided. The distal ends of the GSV tributaries were either clipped or ligated using 3-0 silk

Risk Factors for Chronic SN Following CABG Surgery  3

ligatures (PERMA-HANDTM, Ethicon Endo-SurgeryTM, Inc., Cincinnati, OH, U.S.A.). The subcutaneous tissue and muscles were closed with size 0 or 1 coated polyglactin 910 sutures (VICRYLTM, Ethicon EndoSurgeryTM, Inc.) using a single- or double-layer technique. The skin was closed using size 3-0 polyethylene terephthalate sutures (ETHIBOND EXCELTM, Ethicon Endo-SurgeryTM, Inc.). After closure of the wound, limbs were compressed with a cr^epe bandage, which was replaced after 3 days with an elastic stocking worn for 6 weeks. Postoperative Assessment Patients were assessed 7 days after surgery by a clinician other than those involved in operative care. Afterward, patients were asked to attend the clinic for periodical assessment every 4 weeks for an additional period of 12 weeks. Each assessment session comprised history taking and clinical examination of the limbs where the GSV was harvested. To avoid suggestive questions, other symptoms related to the cardiovascular, respiratory, digestive, and urinary systems, as well as the general functional status of the patient, were initially inquired about. Thereafter, the patients were specifically asked about any pain or sensory abnormality in the lower limbs. To rule out other etiologies contributing to the leg pain, possible causes such as history of leg trauma or strenuous exercise were also inquired about. Patients reporting pain were asked to score both the maximal and the average levels of pain they experienced in the preceding 7 days on an 11-point numerical rating scale (NRS) with 0 representing “no pain” and 10 representing “worst pain possible”. Worst pain scores of “4” or more were denoted as “clinically significant pain”. The nature of pain was then explored inquiring about the patients’ experiencing of somatosensory symptoms such as numbness, pins and needles, burning sensation, or stabbing pain that denoted neuropathic pain, or ache that denoted non-neuropathic pain.16 Patients reporting neuropathic-type pain or abnormal sensations were also asked to outline the affected area using a skin marker. The area of the outlined skin was determined using a transparent plot paper divided into larger squares each of 1-cm2 area and smaller 1-mm2 squares. Eventually, patients were asked if they thought the pain affected their daily activity. The response was noted simply as “yes” or “no”. After history taking, sensation in the lower limbs was examined by observing the response to

pin prick and to stimulation with a piece of cotton wool. To rule out radicular lesions, sensory assessment proceeded systematically by examining the dermatomes supplying each lower limb from proximal distally. Both negative (eg, hypoesthesia or anesthesia) and positive (eg, hyperalgesia or allodynia) somatosensory signs were observed. The examiner using a skin marker with a color different from that used by the patient outlined areas of abnormal sensation. Patients were labeled as suffering from chronic SN if there was sufficient level of certainty to identify the pain as “probable neuropathic pain”.15 This required that the pain had a distinct topographic distribution consistent with the neuroanatomical terriory subserved by the saphenous nerve, had a plausible temporal relation to the surgical trauma, and had concordant negative or positive somatosensory signs on neurological examination. Besides, the pain should have been consistently experienced throughout the follow-up period and patients believed that it affected their daily activity. Management of Leg Pain Patients complaining of pain during the follow-up period were managed conservatively as per the institution’s protocol. For predominantly neuropathic pain, titrated doses of gabapentin were prescribed, increasing the dose every 5 to 7 days as tolerated up to a maximum daily dose of 3,600 mg. For mixed-type pain, paracetamol and ibuprofen were added and the dosage was increased as required to maximum daily doses of 4 g and 2,400 mg, respectively. Statistical Analysis Estimation of the required sample size was based on the expected incidence of the primary outcome measure, the expected dropout rate among recruited subjects, and the number of potential predictors to be included in multivariable analysis. The primary outcome measure was the incidence of moderate-to-severe leg pain in the territory supplied by the saphenous nerve that lasted for longer than 12 weeks after surgery. A previous study by Bruce et al.12 indicated that the incidence of chronic leg pain following saphenous vein harvesting for CABG surgery was around 27% and that approximately 15% of patients reported that the pain was moderate or severe. From previous surveys conducted at our institute, the expected dropout rate was approximately 20%. On the

4 

HAKIM AND NAROUZE

other hand, considering 8 potential predictors to be included in multivariable logistic regression, it was estimated that 80 events had to be observed to avoid an over-fitted model.17 Consequently, it was projected that 668 patients had to be recruited to obtain 534 responders completing the desired follow-up period (expected dropout rate, 20%) and to observe 80 events (expected incidence, 15%). Statistical analysis was performed using MEDCALC© version 12.5 (MedCalc© Software bvba, Ostend, Belgium) and IBM© SPSS© Statistics version 21 (IBM© Corp., Armonk, NY, U.S.A.). The D’Agostino-Pearson test was used to examine the normality of numerical data distribution. Skewed numerical data were presented as median and interquartile range (IQR), and categorical data as number and percentage. Simple binary logistic regression was performed to examine the association between relevant risk factors and the incidence of chronic SN. Variables fulfilling a relaxed threshold criterion of P-value < 0.2 by simple logistic regression and those substantively projected to be candidate predictors for chronic SN, despite not fulfilling this statistical threshold, were entered in multivariable binary logistic regression analysis. A threshold of P-value < 0.2 was selected based on the recommendation that the employment of a conventional level of significance of P-value < 0.05 would be likely to discard pertinent predictors, while more liberal thresholds could result in the inclusion of irrelevant ones.18 All selected explanatory variables were forced into the regression model using the simultaneous (enter) method so as to avoid the indiscriminate elimination of pertinent predictors from the model. The overall model fit was assessed using the 2 log likelihood test, the Hosmer and Lemeshow goodness of fit test, and the correct classification rate. To examine the capability of the model to discriminate between patients with or without the outcome of interest, a receiver operating characteristic (ROC) curve was plotted using the predicted probability as estimated from the full model. All reported P-values were based on two-sided tests of hypothesis. P-value < 0.05 was considered statistically significant.

RESULTS During the study period, 1,167 patients underwent CABG procedures, 991 (84.9%) of which involved the use of GSV grafts. Of the 991 patients who underwent CABG surgeries using GSV grafts, 320 (32.3%) were

primarily excluded because of ineligibility (n = 134, 13.5%) or refusal to participate (n = 186, 18.8%). The reasons for ineligibility were history of surgery on the lower limbs (n = 19, 1.9%), previous stroke involving the lower limbs (n = 21; 2.1%), or involvement of the lower limbs by peripheral vascular disease (n = 41, 4.1%), diabetic neuropathy (n = 36, 3.6%), or radicular pain (n = 17, 1.7%). Six hundred and seventy-one patients (67.7%) were enrolled in the study. Of those enrolled, 526 patients (78.4%) were followed up for the planned period, 109 (16.2%) were lost to follow-up, 28 (4.2%) requested to withdraw from the study, and 8 (1.2%) died during the postoperative period. Of the 526 patients who were followed up for the planned period, 81 (15.4%) consistently had “probable neuropathic pain” of clinically significant severity throughout the follow-up period and were labeled as suffering from “chronic SN”. Fifty-six (10.6%) patients complained of pain that persisted until the end of the follow-up period but was graded either as “unlikely” or as “unconfirmed (possible) neuropathic pain”. Three hundred and eighty-nine (74.0%) patients either had no pain beyond the early postoperative period or complained of pain that was graded as “unlikely to be neuropathic” and resolved by the end of the follow-up period (Figure 1). Table 1 shows the extent of the painful area and severity of pain by 13 weeks after surgery in patients ultimately labeled as having “chronic SN”. The median (IQR) area was 49 (36–69) cm2. The median (IQR) worst and average pain scores were 6 (5–7) and 3 (3–4), respectively. Pain was associated with hyperalgesia in 79 (97.5%) patients and with anesthesia or hypoesthesia in 51 (63.0%) patients. Allodynia and dysesthesia, could be elicited in 41 (50.6%) and 53 (65.4%) patients, respectively. The median (IQR) area with somatosensory signs was 67 (50.5–87) cm2. The results of simple binary logistic regression analysis are shown in Tables 2 and 3. There was statistically significant association between the incidence of chronic SN and the following variables: age (OR, 0.91; 95% CI, 0.88–0.94; P-value, < 0.0001), female gender (OR, 3.23; 95% CI, 1.16–1.31; P-value, < 0.0001), body mass index (BMI) (OR, 1.24; 95% CI, 1.16–1.31; P-value, < 0.0001), diabetes mellitus (OR, 2.14; 95% CI, 1.28–3.58; P-value, 0.004), dissection of the GSV from distal proximally (OR, 5.65; 95% CI, 3.23–9.87; P-value, < 0.0001), and closure of the leg wound in two layers (OR, 3.35; 95% CI, 2.06–5.44; P-value, < 0.0001).

Risk Factors for Chronic SN Following CABG Surgery  5

Figure 1. Flow chart showing patient enrollment, follow-up, and analysis.

The results of multivariable binary logistic regression are shown in Table 4. Of the included variables, only age (OR, 0.92; 0.88–0.95; P-value, < 0.0001), female gender (OR, 2.28; 95% CI, 1.21–4.29; P-value, 0.011), BMI (OR, 1.25; 95% CI, 1.17–1.35; P-value, < 0.0001),

diabetes mellitus (OR, 2.13; 95% CI, 1.13–4.01; P-value, 0.020), distal-to-proximal dissection of the GSV (OR, 7.28; 95% CI, 3.62–14.66; P-value, < 0.0001), and wound closure in two layers (OR, 3.28; 95% CI, 1.81–5.95; P-value, 0.0001) were independent

6 

HAKIM AND NAROUZE

predictors for chronic SN. The full model had an adequate overall fit as evidenced by a 2 log likelihood of 306.2 (null model-2 log likelihood, 451.9; Χ2, 145.7; DF, 8; P-value, < 0.0001) and a statistically nonsignifTable 1. Extent of Painful Area and Severity of Pain by 13 Weeks after Surgery in Patients Ultimately Labeled as Having Chronic Saphenous Neuralgia Variable

Metric

Number of patients developing chronic SN Painful area as outlined by patient, cm2 Worst pain score in preceding 7 days Average pain score in preceding 7 days Associated somatosensory signs Hyperalgesia Allodynia Dysesthesia Anesthesia/hypoesthesia Area with abnormal sensation as assessed by examiner, cm2

81 (15.4%) 49 (36–69) 6 (5–7) 3 (3–4) 79 (97.5%) 41 (50.6%) 53 (65.4%) 51 (63.0%) 67 (51–87)

Data are presented as number (%) or median (IQR).

icant Hosmer and Lemeshow test (Χ2, 4.1; DF, 8; P-value, 0.849). The model could classify 87.8% of the patients correctly. The receiver operating characteristic (ROC) curve derived from the model is shown in Figure 2. The area under the ROC curve (AUC) was 0.870 (95% CI, 0.838–0.898; P-value, < 0.0001). The best cutoff criterion was a predicted probability of > 0.131 (Youden J index, 0.59). This had a sensitivity of 82.7% (95% CI, 72.7% to 90.2%) and a specificity of 76.2% (95% CI, 71.9% to 80.1%).

DISCUSSION The present study showed that the incidence of chronic SN, defined as persistence of probable neuropathic pain of clinically significant severity in the territory supplied by the saphenous nerve beyond 12 weeks

Table 2. Patients’ Characteristics Variable

All Patients (n = 526)

Chronic SN (n = 81)

No Chronic SN (n = 445)

Age, yr Female gender BMI, kg/m2 Current smoker Diabetes mellitus Hypertension Atrial fibrillation Dylipidemia Renal impairment History of ACS Previous cardiac surgery History of TIA Preoperative EF ≤ 0.4 NYHA class III/IV

60.6 (7.7) 111 (21.1%) 23.9 (3.8) 353 (67.1%) 116 (22.1%) 395 (75.1%) 51 (9.7%) 427 (81.2%) 48 (9.1%) 164 (31.2%) 58 (11.0%) 22 (4.2%) 159 (30.2%) 289 (54.9%)

56.0 (7.4) 33 (40.7%) 27.0 (5.1) 53 (65.4%) 28 (34.6%) 63 (77.8%) 9 (11.1%) 62 (76.5%) 9 (11.1%) 26 (32.1%) 12 (14.8%) 5 (6.2%) 27 (33.3%) 43 (53.1%)

61.4 (7.5) 78 (17.5%) 23.3 (3.2) 300 (67.4%) 88 (19.8%) 332 (74.6%) 42 (9.4%) 365 (82.0%) 39 (8.8%) 138 (31.0%) 46 (10.3%) 17 (3.8%) 132 (29.7%) 246 (55.3%)

Odds Ratio

95% CI for Odds Ratio

P-Value

0.91 3.23 1.24 0.91 2.14 1.19 1.20 0.72 1.30 1.05 1.51 1.66 1.19 0.92

0.88–0.94 1.95–5.37 1.16–1.31 0.56–1.51 1.28–3.58 0.68–2.10 0.56–2.57 0.41–1.26 0.60–2.80 0.63–1.75 0.76–2.99 0.59–4.62 0.72–1.96 0.57–1.47

< 0.0001 < 0.0001 < 0.0001 0.727 0.004 0.544 0.640 0.248 0.501 0.846 0.239 0.335 0.509 0.715

Data are presented as mean (SD) or number (percentage). 95% CI, 95% confidence interval; ACS, acute coronary syndrome; BMI, body mass index; EF, ejection fraction; NYHA, New York heart association; SN, saphenous neuralgia; TIA, transient ischemic attacks.

Table 3. Surgical Details Variable CABG procedure Valve replacement/repair done in addition to CABG CABG done as emergency procedure Three or four coronary vessels grafted CPB time > 120 min Aortic cross-clamping time > 75 min Hypothermia on CPB < 30°C Surgical time > 240 min Harvesting of GSV graft Leg incision extending above knee Distal-to-proximal dissection of the GSV Closure of leg wound in two layers Continuous suturing technique for closure of leg wound

All Patients (n = 526)

Chronic SN (n = 81)

No Chronic SN (n = 445)

74 (14.1%)

12 (14.8%)

85 (16.2%) 484 (92%) 76 (14.4%) 85 (16.2%) 63 (12.0%) 79 (15.0%) 173 (32.9%) 69 (13.1%) 166 (31.6%) 312 (59.3%)

Odds Ratio

95% CI for Odds Ratio

62 (13.9%)

1.07

0.55–2.10

0.834

15 (18.5%) 73 (90.1%) 11 (13.6%) 12 (14.8%) 8 (9.9%) 13 (16.0%)

70 (15.7%) 411 (92.4%) 65 (14.6%) 73 (16.4%) 55 (12.4%) 66 (14.8%)

1.22 0.75 0.92 0.89 0.78 1.10

0.66–2.25 0.34–1.70 0.46–1.83 0.46–1.72 0.36–1.70 0.57–2.10

0.531 0.496 0.809 0.721 0.528 0.778

27 29 45 47

146 (32.8%) 40 (9.0%) 121 (27.2%) 265 (59.6%)

1.02 5.65 3.35 0.94

0.62–1.69 3.23–9.87 2.06–5.44 0.58–1.52

0.926 < 0.0001 < 0.0001 0.797

(33.3%) (35.8%) (55.6%) (58.0%)

P-Value

Data are presented as number (percentage). 95% CI, 95% confidence interval; CABG, coronary artery bypass graft; CPB, cardiopulmonary bypass; GSV, great saphenous vein.

Risk Factors for Chronic SN Following CABG Surgery  7

Table 4. Multivariable Binary Logistic Regression Model for Determinants of Chronic Saphenous Neuralgia Variable

B

SE

Age, yr Female gender BMI, kg/m2 Diabetes mellitus Leg incision extending above knee Distal-to-proximal dissection of the GSV Closure of leg wound in two layers Continuous suturing technique for closure of leg wound Constant

0.09 0.82 0.22 0.76 0.20 1.99 1.19 0.07 3.32

0.02 0.32 0.04 0.32 0.32 0.36 0.30 0.30 1.48

Wald Χ2

DF

Odds Ratio

95% CI for Odds Ratio

P-Value

18.99 6.54 37.64 5.46 0.39 30.96 15.30 0.05 5.06

1 1 1 1 1 1 1 1 1

0.92 2.28 1.25 2.13 0.82 7.28 3.28 0.94 0.04

0.88–0.95 1.21–4.29 1.17–1.35 1.13–4.01 0.44–1.53 3.62–14.66 1.81–5.95 0.52–1.70

< 0.0001 0.011 < 0.0001 0.020 0.534 < 0.0001 0.0001 0.828 0.024

95% CI, 95% confidence interval; B, regression coefficient; BMI, body mass index; DF, degree of freedom; GSV, great saphenous vein; SE, standard error; Wald Χ2, Wald chi square statistic.

Figure 2. Receiver operating characteristic (ROC) curve derived from the multivariable binary logistic regression model. Dotted lines represent 95% confidence bounds (95% CI). Area under the curve (AUC), 0.870 (95% CI, 0.838–0.898; P-value, < 0.0001). The best cutoff criterion is a predicted probability of > 0.131 Youden J index 0.59). This has a sensitivity of 82.7% (95% CI, 72.7% to 90.2%) and a specificity of 76.2% (95% CI, 71.9% to 80.1%).

following harvesting of the GSV for GABG surgery, was approximately 15.4%. Younger age, female gender, higher BMI, diabetes mellitus, distal-to-proximal dissection of the GSV, and closure of the leg wound in two layers were identified in this trial as independent risk factors. Chronic leg pain following GSV harvesting is common, occurring in up to 27% of patients.12 Although pain is described as mild in approximately one-third of such patients, the overall incidence of moderate and severe chronic leg pain after GSV harvesting has been

reported to be on the order of 15%,12 a figure close to that observed in the present study. The exact etiology of chronic SN after GSV harvesting for CABG surgery is not fully understood. However, both patient and surgical factors have been implicated. For instance, compression of the saphenous nerve or its branches has been proposed as one possible mechanism for SN. In this regard, Nair et al.4 reported that closure of the wound in two layers was associated with higher incidence of SN. The authors projected that while closure of the subcutaneous tissues as a separate layer may offer the privilege of better approximation of the wound edges and more efficient hemostasis, it could predispose the nerve fibers to undue compression, especially with the development of tissue edema in the postoperative period. In fact, the present study demonstrated that the use of a double-layer technique for wound closure was an independent risk factor for chronic SN. Another independent risk factor demonstrated by the present study was the dissection of the saphenous veins in an upward (ie, from the medial malleolus upwards) rather than a downward (ie, from the saphenofemoral junction downward) direction. Related findings were also reported by Ramasastry et al.13 who demonstrated that the upward dissection technique was associated with significantly higher incidence of sensory abnormality in the territory supplied by the saphenous nerve at 3 months after surgery. The authors correlated their findings with cadaveric examination which revealed that dissection of the GSV in an upward direction could result in avulsion of the pretibial and infrapatellar branches of the saphenous nerve, a complication that was not observed with the downward dissection technique, and recommended that the GSV be dissected in a downward direction to minimize damage to the saphenous nerve and its branches. Another cadaveric study by Holme et al.19 ended up with a similar conclusion.

8 

HAKIM AND NAROUZE

Of the patient characteristics demonstrated to be independent risk factors for chronic SN in the present study were younger age, female gender, higher BMI, and diabetes mellitus. The relation between age and chronic postsurgical pain (CPSP) was demonstrated by previous investigators following cardiac12,14 and noncardiac20,21 procedures. Although there is presently no objective explanation for this phenomenon, it has been suggested that younger patients could be less likely to accept pain or be more likely to report pain symptoms compared with older patients who are often satisfied that they have survived a cardiac surgery despite their experiencing pain.22 Another patient-related risk factor demonstrated in the present study was BMI. Higher BMI was previously shown to be a risk factor for CPSP following cardiac surgery.12 This applied for both chronic chest pain and chronic leg pain. A similar relation between BMI and CPSP has been described following other surgeries such as inguinal hernia repair21 and mastectomy20. A possible explanation for this association might be that peripheral nerves, especially the smaller main nerves or their branches, would be more difficult to identify in the obese, who would thus be more vulnerable to trauma of these structures during surgical dissection. In concordance with a previous trial that reported that diabetes mellitus was a risk factor for postoperative wound sensitivity disorders following radial artery procurement for CABG surgery,23 the present study showed that diabetes mellitus was an independent predictor for chronic SN. Nonetheless, another trial failed to establish any relation between diabetes mellitus and wound sensitivity disorders for either upper or lower limb harvesting sites.14 The relation between gender and development of chronic pain disorders after surgery has been inconsistent in the literature. For instance, a couple of studies reported no difference between males and females in regard to the prevalence of CPSP.24,25 In contrast to these reports, there is evidence from the setting of hernia repair that female patients could be at more risk for CPSP.26–28 In agreement with these reports,26–28 the present study demonstrated that female gender was an independent risk factor for chronic SN following harvesting of the GSV. However, to the authors’ knowledge, no such relation has been described previously in this setting. The term “saphenous neuralgia” or “neuropathy” has been used to describe a heterogeneous assortment of neurological symptoms, including pain, that develop

after injury to the saphenous nerve or its branches has been sustained.4–8,11 Although there is a propensity for neuralgic symptoms to improve with time,4 these may be perpetuated for a fairly prolonged period in a proportion of patients with unfavorable effects on their quality of life, particularly if pain is a conspicuous element of their symptomatology.12 Lacking a consensus definition for chronic SN, we have defined the condition as persistence of probable neuropathic pain of moderate or severe intensity in the saphenous nerve territory for longer than 12 weeks, a time frame that is usually employed for definition of chronic pain syndromes.29,30 Although other investigators regarded the perpetuation of pain of any degree beyond this time interval as persistent postsurgical pain (PPSP),30 the current authors were concerned with clinically significant pain defined arbitrarily as a worst pain score of 4 or more on an 11-point NRS which was supposed to impact a patient’s quality of life. This definition has been employed previously by at least one trial conducted in a different clinical setting.31 In compliance with the recommendations for trials involving chronic pain, the present study employed an 11-point rating scale to score pain in the preceding 7 days.32 The present study used a certainty grade of “probable neuropathic pain” and employed none of the known screening tools or diagnostic tests to work out a higher certainty grade of “definite neuropathic pain”. Although this may be regarded as a limitation, we have not applied these tools for the following reasons. First, current guidelines recommend that screening tools such as the LANSS Pain Scale33 or the NP4 (DN4) test34 should not be regarded as a substitute for a judicious clinical examination, as they may have a false negative rate in the order of 10% to 20%.35 Second, the current recommendation is that methods such as quantitative sensory testing (QST) may accompany bedside testing to document the sensory abnormality. However, as abnormalities may be elicited in non-neuropathic pain, QST cannot provide sufficient evidence for making the diagnosis of neuropathic pain.35 Likewise, neurophysiologic tools such as standard nerve conduction studies may not provide adequate information on small fiber function.35 Third, as diagnostic tools such as neuroimaging and laboratory testing would provide no conclusive evidence for involvement of the saphenous nerve by a disease process, it was deemed unfeasible to subject the patients to further testing to obtain a level of “definite neuropathic pain”. Moreover, as per the NeuPSIG criteria, the diagnosis may be equally

Risk Factors for Chronic SN Following CABG Surgery  9

established upon fulfillment of a certainty level of either “definite” or “probable” neuropathic pain on the proposed grading system.15 Although the primary objective for the present study was to identify risk factors for chronic SN rather than to develop a prediction rule, the model obtained in the present study had good predictive value as evidenced by an area under the ROC curve of approximately 0.88 and a correct classification rate of approximately 88%. However, one limitation to the present trial is that the model was not validated using either a training sample or another validation cohort. This could have biased the efficiency of the model as regression models are inherently more fitted for the data from which they are derived.36,37 Therefore, validation of the proposed model on another cohort is strongly recommended. Another limitation to the present trial is the considerably high proportion of patients who were primarily excluded, lost to follow-up, or withdrew from the study, which could have also biased our results. In accordance with the definition of chronic pain adopted by the current trial,29,30 patients were followed up for approximately 3 months after surgery. Extrapolation of the results to longer periods should therefore be exercised with caution. Summary Chronic SN after CABG surgery utilizing GSV grafts is not uncommon. Risk factors identified in this study are younger age, female gender, higher BMI, diabetes mellitus, dissection of the saphenous veins in a distalto-proximal direction, and closure of the leg wound in 2 layers. These results need to be validated by further trials using other cohorts. A relevant implication for the current findings may be that attention to the surgical technique and, when feasible, weight reduction prior to surgery, could reduce the risk of chronic SN. These assumptions may serve as a plausible hypothesis to be tested by future researchers.

REFERENCES 1. Epstein AJ, Polsky D, Yang F, Yang L, Groeneveld PW. Coronary revascularization trends in the United States, 2001– 2008. JAMA. 2011;305:1769–1776. 2. Goldman S, Zadina K, Moritz T, et al. VA Cooperative Study Group. Long-term patency of saphenous vein and left internal mammary artery grafts after coronary artery bypass surgery: results from a Department of Veterans Affairs Cooperative Study. J Am Coll Cardiol. 2004;44:2149–2156.

3. Verma S, Szmitko PE, Weisel RD, et al. Should radial arteries be used routinely for coronary artery bypass grafting? Circulation. 2004;110:e40–e46. 4. Nair UR, Griffiths G, Lawson RA. Postoperative neuralgia in the leg after saphenous vein coronary artery bypass graft: a prospective study. Thorax. 1988;43: 41–43. 5. Mountney J, Wilkinson GA. Saphenous neuralgia after coronary artery bypass grafting. Eur J Cardiothorac Surg. 1999;16:440–443. 6. Defalque RJ, McDanal JT. Proximal saphenous neuralgia after coronary artery bypass graft. Anesth Analg. 1995;80:620–621. 7. Schnall B, San Luis F. Saphenous neuropathy following coronary bypass. Orthopedic Rev. 1979;8:121–122. 8. Chauhan BM, Kim DI, Wainanel SF. Sauhenous neurouathv following coronary artery bypass surgery. NY State J Med. 1981;81:222–223. 9. Mercer D, Morrell NT, Fitzpatrick J, et al. The course of the distal saphenous nerve: a cadaveric investigation and clinical implications. Iowa Orthop J. 2011;31:231–235. 10. Dayan V, Cura L, Cubas S, Carriquiry G. Surgical anatomy of the saphenous nerve. Ann Thorac Surg. 2008;85:896–900. 11. Adar R, Meyer E, Zweig A. Saphenous neuralgia: a complication of vascular reconstructions below the inguinal ligament. Ann Surg. 1979;190:609–613. 12. Bruce J, Drury N, Poobalan AS, Jeffrey RR, Smith WC, Chambers WA. The prevalence of chronic chest and leg pain following cardiac surgery: a historical cohort study. Pain. 2003;104:265–273. Erratum in Pain. 2004;112:413. 13. Ramasastry SS, Gregory OD, Futrell JW. Anatomy of the saphenous nerve: relevance to the saphenous vein stripping. Am Surgeon. 1987;53:274–277. 14. Dick F, Hristic A, Roost-Kr€ ahenb€ uhl E, et al. Persistent sensitivity disorders at the radial artery and saphenous vein graft harvest sites: a neglected side effect of coronary artery bypass grafting procedures. Eur J Cardiothorac Surg. 2011;40:221–226. 15. Treede RD, Jensen TS, Campbell JN, et al. Neuropathic pain: redefinition and a grading system for clinical and research purposes. Neurology. 2008;70:1630–1635. 16. Boureau F, Doubrere JF, Luu M. Study of verbal description in neuropathic pain. Pain. 1990;42: 145–152. 17. Vittinghoff E, Shiboski SC, Glidden DV, McCulloch CE. Regression Methods in Biostatistics: Linear, Logistic, Survival, and Repeated Measures Models. 1st ed.. New York: Springer; 2004. 18. Mickey J, Greenland S. A study of the impact of confounder selection criteria on effect estimation. Am J Epidemiol. 1989;129:125–137. 19. Holme JB, Holme K, Sørensen LS. The anatomic relationship between the long saphenous vein and the saphenous nerve. Relevance for radical varicose vein surgery. Acta Chir Scand. 1988;154:631–633.

10 

HAKIM AND NAROUZE

20. Smith WCS, Bourne D, Squair J, Phillips DO, Chambers WA. A retrospective cohort study of post mastectomy pain syndrome. Pain. 1999;83:91–95. 21. Poobalan AS, Bruce J, King PM, Chambers WA, Krukowski ZH, Smith WCS. Chronic pain and quality of life following open inguinal hernia repair. Br J Surg. 2001; 88:1122–1126. 22. Greene MA, Malias MA. Arm complications after radial artery procurement for coronary bypass operation. Ann Thorac Surg. 2001;72:126–128. 23. Budillon AM, Nicolini F, Agostinelli A, et al. Complications after radial artery harvesting for coronary artery bypass grafting: our experience. Surgery. 2003;133: 283–287. 24. Cho SH, Kim YR, Lee JH, et al. A questionnaire study investigating the prevalence of chronic postoperative pain. Korean J Anesthesiol. 2012;62:40–46. 25. Liem MS, van Duyn EB, van der Graaf Y, van Vroonhoven TJ, Coala Trial Group. Recurrences after conventional anterior and laparoscopic inguinal hernia repair: a randomized comparison. Ann Surg. 2003;237:136–141. 26. Bay-Nielsen M, Perkins FM, Kehlet H. Pain and functional impairment 1 year after inguinal herniorrhaphy: a nationwide questionnaire study. Ann Surg. 2001;233:1–7. 27. Mori T, Souda S, Nezu R, Yoshikawa Y. Results of performing mesh plug repair for groin hernias. Surg Today. 2001;31:129–132. 28. Søndenaa K, Nesvik I, Breivik K, Kørner H. Long-term follow-up of 1059 consecutive primary and recurrent inguinal hernias in a teaching hospital. Eur J Surg. 2001;167:125–129.

29. Merskey H, Bogduk H. Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms. Seattle: IASP Press; 1994. 30. Johansen A, Romundstad L, Nielsen CS, Schirmer H, Stubhaug A. Persistent postsurgical pain in a general population: prevalence and predictors in the Tromsø study. Pain. 2012;153:1390–1396. 31. Hakim SM, Narouze S, Shaker NN, Mahran MA. Risk factors for new-onset persistent low-back pain following nonobstetric surgery performed with epidural anesthesia. Reg Anesth Pain Med. 2012;37:175–182. 32. Dworkin RH, Turk DC, Farrar JT, et al. IMMPACT. Core outcome measures for chronic pain clinical trials: IMMPACT recommendations. Pain. 2005;113:9–19. 33. Bennett M. The LANSS Pain Scale: the Leeds assessment of neuropathic symptoms and signs. Pain. 2001;92:147– 157. 34. Bouhassira D, Attal N, Alchaar H, et al. Comparison of pain syndromes associated with nervous or somatic lesions and development of a new neuropathic pain diagnostic questionnaire (DN4). Pain. 2005;114:29–36. 35. Cruccu G, Sommer C, Anand P, et al. EFNS guidelines on neuropathic pain assessment: revised 2009. Eur J Neurol. 2010;17:1010–1018. 36. Wasson JH, Sox HC, Neff RK, Goldman L. Clinical prediction rules. Applications and methodological standards. N Engl J Med. 1985;313:793–799. 37. Concato J, Feinstein AR, Holford TR. The risk of determining risk with multivariable models. Ann Intern Med. 1993;118:201–210.

Risk Factors for Chronic Saphenous Neuralgia Following Coronary Artery Bypass Graft Surgery Utilizing Saphenous Vein Grafts.

The aim of this trial was to determine risk factors for chronic saphenous neuralgia (SN) following harvesting of the great saphenous vein (GSV) for co...
198KB Sizes 1 Downloads 14 Views