This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Iohom, G. Articles by Shorten, G. D. Search for Related Content PubMed PubMed Citation Articles by Iohom, G. Articles by Shorten, G. D. Related Collections Mechanisms Pain Pharmacology

---

ANALGESIA

The Associations Between Severity of Early Postoperative Pain, Chronic Postsurgical Pain and Plasma Concentration of Stable Nitric Oxide Products After Breast Surgery

Gabriella Iohom, FCARCSI^*, Hamza Abdalla, MRCP^*, James O'Brien, FCARCSI^*, Szilvia Szarvas, MD, DEEA^*, Vivienne Larney, FCARCSI^*, Elisabeth Buckley, FCARCSI^*, Mark Butler, MSc, and George Declan Shorten, PhD^*

*Department of Anaesthesia and Intensive Care Medicine, Department of Clinical Biochemistry, Cork University Hospital, Cork, Ireland.

Address correspondence and reprint requests to George Declan Shorten, PhD, Department of Anesthesia & Intensive Care Medicine, Cork University Hospital, Cork, Ireland. Address e-mail to shorteng{at}shb.i.e.

Abstract

In this study, we compared the effects of two analgesic regimens on perioperative nitric oxide index (NOx) and the likelihood of subsequent development of chronic postsurgical pain (CPSP) after breast surgery and sought to determine the association among early postoperative pain, NOx, and the likelihood of subsequent development of CPSP. Twenty-nine consecutive ASA I or II patients undergoing breast surgery with axillary clearance were randomly allocated to one of two groups. Patients in group S (n = 15) received a standard intraoperative and postoperative analgesic regimen (morphine sulfate, diclofenac, dextropropoxyphene hydrochloride + acetaminophen prn). Patients in group N (n = 14) received a continuous paravertebral block (for 48 h) and acetaminophen and parecoxib (followed by celecoxib up to 5 days). Visual analog scale pain scores at rest and on arm movement were recorded regularly until the fifth postoperative day. A telephone interview was conducted 10 wk postoperatively. The McGill Pain Questionnaire was used to characterize pain. NOx was estimated preoperatively, at the end of surgery, 30 min and 2, 4, 12, 24, 48 h postoperatively. Twelve (80%) patients in group S and no patient in group N developed CPSP (P = 0.009). Compared with patients with a pain rating index 1 (n = 18) 10 wk postoperatively, patients with a pain rating index = 0 (n = 11) had lesser visual analog scale pain scores on movement at each postoperative time point from 30 min until 96 h postoperatively (P < 0.005) and at rest 30 min (0.6 ± 1.5 versus 30.2 ± 26.8; P = 0.004), 4 h (2.3 ± 7.5 versus 19.0 ± 25.8; P = 0.013), 8 h (4.4 ± 10.2 versus 21.4 ± 27.0; P = 0.03) and 12 h (0.7 ± 1.2 versus 15.4 ± 27.0; P = 0.035) postoperatively. NOx values were greater in group N compared with group S 48 h postoperatively (40.6 ± 20.1 versus 26.4 ± 13.5; P = 0.04).

Treatment of postoperative pain is important for humanitarian and ethical reasons, but it also has economic implications, as pain may contribute to prolonged convalescence (1). It is also likely that adequate postoperative analgesia leads to improved patient outcome (2).

In some patients, postsurgical pain persists long after natural healing processes have been completed. Chronic postsurgical pain (CPSP) is defined as pain developed after a surgical procedure, of at least 2 mo duration, in which other causes (i.e., malignancy or chronic infection) have been excluded, the possibility of pain continuing from a preexisting condition explored and exclusion attempted (3).

CPSP after breast surgery is common, with a variable incidence (4,5). Forty-nine percent of patients who undergo mastectomy with reconstruction have pain 1 yr later, compared with 31% of those undergoing mastectomy alone and 22% of those undergoing breast reduction (4). Of 126 women who underwent axillary dissection (but neither radiotherapy nor chemotherapy), 70% complained of numbness, 33% of pain, 25% of arm weakness, 24% of arm swelling, and 15% of stiffness 1 yr after surgery (5).

Early postoperative pain is the only factor that significantly predicts long-term pain after thoracotomy (6). Similarly, chronic pain after open inguinal hernia repair can be predicted by the intensity of early postoperative pain (7).

Nitric oxide (NO) plays an important role in nociception processing. Nociceptive afferent activation (e.g., peripheral nerve injury) results in increased excitability of spinal neurons, a phenomenon known as central sensitization. Central sensitization is at least partially mediated by activation of N-methyl-d-aspartate receptors, which could lead ultimately to NO production (8). Although the link between local effects and systemic NO products is not readily apparent, there is evidence of a correlation between the two (9).

The objective of this prospective randomized study was: a) to compare the effects of two analgesic regimens on perioperative NO levels and the likelihood of subsequent development of CPSP after breast surgery with axillary node clearance and b) to determine the association between early postoperative pain, perioperative plasma NO concentrations, and the likelihood of subsequent development of CPSP.

METHODS

With Institutional Ethics Committee approval and having obtained written informed consent from each patient, 29 ASA I or II patients undergoing breast surgery (mastectomy or breast tumor resection) with axillary node clearance were randomized using a random numbers table to one of two groups (S and N). Patients in group S received a standard analgesic regimen (morphine sulfate 0.1 mg/kg IM 4 hourly prn, diclofenac 100 mg pr 12 hourly prn, dextropropoxyphene hydrochloride 32.5 mg + acetaminophen 325 mg x 2 po 6 hourly prn). Those in group N received an aggressive "around the clock" analgesic regimen: parecoxib 40 mg IV 12 hourly (a selective cyclooxygenase (COX)-2 inhibitor, commenced 12 h preoperatively) for as long as the IV cannula was in place, then commenced on celecoxib 200 mg po 12 hourly until postoperative day 5, acetaminophen 1 g po 6 hourly and a continuous paravertebral block using bupivacaine 0.25% (10 mL 12 hourly) up to 48 h postoperatively. Rescue analgesia consisted of tramadol 100 mg IM or po.

Patients received no preanesthetic medication. Patients assigned to group N had a paravertebral catheter inserted while awake in the sitting position. A 25-gauge needle was inserted 2.5 cm lateral (on the operative side) from the cephalad edge of the third thoracic vertebral spinal process, and the skin, subcutaneous tissue and periosteum of the transverse process were anesthetized with 2–5 mL of lidocaine 1%. The paravertebral block was performed with a Tuohy needle and the loss-of-resistance to air technique was used, seeking contact with the lateral process of the third thoracic vertebra as a landmark before advancing the needle into the paravertebral space. Bupivacaine 0.25% 10 mL was injected into the paravertebral space in small aliquots with repeated aspiration tests before the paravertebral catheter was inserted. Patients in group N received parecoxib 40 mg IV 12 h preoperatively and immediately after induction of anesthesia. Intraoperative analgesia in group S consisted of diclofenac 100 mg pr and morphine sulfate prn IV.

With standard monitoring in place (pulse oximetry, electrocardiography, noninvasive arterial blood pressure, and inspired and end-tidal partial pressures of sevoflurane, N₂O, and O₂ (Datex AS/3 monitor; Datex Corp., Helsinki, Finland), anesthesia was induced by administering sevoflurane 8% in 100% O₂ and maintained using the clinically indicated concentration of sevoflurane in an O₂/N₂O mixture with an Fio₂ of 0.30. Intermittent positive pressure ventilation was used to maintain Petco₂ between 4.0 to 4.5 kPa. Muscle relaxation was achieved with 0.1 mg/kg vecuronium and incremental doses as necessary, monitored using a transcutaneous nerve stimulator (MiniStim MS-IIIA; Life-Tech, Inc., Houston, TX) and reversed with neostigmine 50 µg/kg combined with 10 µg/kg of glycopyrrolate.

The following assessments were performed by a dedicated investigator (HA) unaware of the patients' group assignment. The Hospital Anxiety and Depression (HAD) Scale was applied preoperatively, before discharge from hospital on postoperative day 5, and 10 wk after surgery (10). Visual analog scale (VAS) pain scores (0 = no pain and 100 = worst pain imaginable) at rest and on arm movement (abduction and extension) were recorded at 2, 4, 8, and 12 h and 2, 3, 4 and 5 days postoperatively. Postoperative nausea and vomiting (PONV) and pruritus were recorded at each of these times. The severity of PONV and pruritus was assessed using a VAS ranging from no symptom (0 mm) to worst imaginable symptoms (100 mm). The VAS scores were categorized as follows: 0 mm = no symptoms; 10–30 mm = mild symptoms; 30–70 mm = moderate symptoms, and 70–100 mm = severe symptoms.

A telephone interview was conducted 2–3 mo postoperatively (at the midperiod of the menstrual cycle) by the same blinded investigator. A standard questionnaire was used (2) (Appendix 1). CPSP was considered to be present if the answer to the question "do you have chronic pain as a result of your breast surgery?" was "yes." The McGill Pain Questionnaire (long form) administered over the phone was used to characterize pain at this time point (11). Two indices are produced from this information: i) the Pain Rating Index (PRI): the sum total of the ranked values of each chosen descriptor and ii) the Present Pain Intensity: a categorical scale using descriptors from no pain to excruciating pain.

Both "yes" versus "no" groups, and PRI 1 versus PRI = 0 were compared in terms of perioperative VAS pain scores and nitric oxide index (NOx). A history of chemotherapy or radiotherapy was also noted.

Plasma concentrations of stable NO products (NOx) were estimated preoperatively (T1), after induction of anesthesia (T2), at the end of surgery before discontinuing sevoflurane (T3), and at 30 min, 2, 4, 12, 24, and 48 h postoperatively (T4–9). NO oxidation products were measured using a Nitric Oxide Chemiluminescent Analyzer, Sievers 280 NOA^TM (Sievers Instruments, Boulder, CO).

Based on an incidence of 50% CPSP after breast surgery (12) and accepting a reduction to 10% of this incidence as clinically significant, a minimum sample size of 14 patients per group was chosen. Patients' weight and height and duration of surgery were compared between groups using two-tailed, unpaired Student's t-test. Repeated-measures analysis of variance followed by Bonferroni's post hoc test was used to compare analgesic consumption, VAS for pain, PONV, pruritus, and NOx between groups. Fisher's exact test was used to compare nonparametric data. P < 0.05 was considered significant.

RESULTS

Groups were similar in terms of demographics and surgery. The number of patients who subsequently underwent radiotherapy and/or chemotherapy was similar in the two groups (Table 1). Duration of surgery was similar in both groups (157 ± 29 and 143 ± 23 min for group S and group N, respectively).

The HAD scores were similar in the 2 groups preoperatively and at 5 days and 10 wk postoperatively (Table 2). These scores were also similar in patients who did and did not subsequently develop CPSP.

Seven patients in group S and three patients in group N needed postoperative rescue analgesia in the form of tramadol. No major complication occurred as a result of the paravertebral blocks. One patient in the aggressive treatment group (group N) presented an ipsilateral Horner's syndrome on arrival to the recovery room that resolved spontaneously.

In group N, VAS pain scores were less compared with group S, at rest 30 min postoperatively (6 ± 11 versus 31 ± 29; P = 0.04) (Fig. 1) and on movement at 30 min, 12 h, and on postoperative days 1, 2, 3, 4, 5 (P < 0.04) (Fig. 2). Although aggressive pain management only modestly improved pain at rest, considerable benefit was obtained on movement. PONV was less 30 min postoperatively in group N compared with group S (11 ± 6 versus 20 ± 10; P = 0.012). Pruritus was less in group N 2 h postoperatively compared with group S (0 ± 0 versus 12 ± 4; P = 0.01).

View larger version (9K): [in this window] [in a new window]   Figure 1. Visual analog scale (VAS) at rest (group S versus group N). Data are mean (sd). *P = 0.04 refers to between-group comparison. VAS = visual analog scale; POD = postoperative day.

View larger version (9K): [in this window] [in a new window]   Figure 2. Visual analog scale (VAS) on movement (group S versus group N). Data are mean (sd). *P < 0.04 refers to between-group comparisons. VAS = visual analog scale; POD = postoperative day.

The incidence of CPSP ("yes" answer as defined in the Methods section) was 80% (12/15 patients) in group S and 0% (0/14) in group N. This yields a ² value with one degree of freedom of 6.78, which is significant at P = 0.009. Although no patient sought active treatment in a formal pain clinic setting, two patients took nonprescription analgesia. Pain interfered with daily activities in seven and with sleep patterns in five patients. Two complained of constant pain, and 10 described the pain as intermittent (lasting for seconds in 4 and for minutes in 6 patients). Pain was experienced in the following locations: arm (seven), scar/breast (eight), and shoulder (two cases).

The McGill pain questionnaire revealed both greater PRI and present pain index (PPI) in group S compared with group N (11.6 ± 17.7 versus 0.4 ± 0.8; P = 0.01 and 0.4 ± 0.8 versus 0 ± 0; P = 0.04, respectively).

All patients in group S had a PRI 1 at 10 wk postoperatively, whereas only 3 of 14 patients (21%) in group N had a PRI >1 (explained by numbness). Compared with patients with a PRI 1 (n = 18), patients with a PRI = 0 (n = 11) had lower VAS pain scores at rest 30 min after surgery (0.6 ± 1.5 versus 30.2 ± 26.8; P = 0.004), 4 h (2.3 ± 7.5 versus 19 ± 25.8; P = 0.013), 8 h (4.4 ± 10.2 versus 21.4 ± 27.0; P = 0.03) and 12 h (0.7 ± 1.2 versus 15.4 ± 27.0; P = 0.035) postoperatively and on movement at each postoperative time point from 30 min until postoperative day 4 (P < 0.005).

Overall NOx values were less 12 h postoperatively compared with baseline (21.0 ± 12.5 versus 29.9 ± 21.6 µmol/L; P = 0.03) (Table 3). Although a tendency to greater NOx values in group N compared with group S persisted throughout the perioperative period, this reached statistical significance only at 48 h postoperatively (40.6 ± 20.1 versus 26.4 ± 13.5 µmol/L; P = 0.04). Perioperative NOx values were similar in patients who did and did not subsequently develop CPSP. No association between NOx and the subsequent development of CPSP could be demonstrated.

DISCUSSION

The most important findings of this study are that i) no patient in group N (aggressive perioperative pain management) developed subsequent CPSP and ii) adequacy of postoperative analgesia is an important determinant of CPSP after breast surgery.

CPSP is common but often under-recognized, neglected, or misdiagnosed. Chronic pain is costly to society in terms of suffering and disability. For humanitarian, medical, and economic reasons, the problem of chronic pain after surgery should be addressed (12). Women who undergo breast surgery experience chest wall, breast, or scar pain (range, 11%–57%), phantom breast pain (13%–24%), and arm and shoulder pain (12%–51%) (12). The incidence of pain in one or more of these sites is approximately 50% 1 year after breast surgery for cancer (12). The 75% incidence of CPSP at 10 weeks postoperatively in our study is slightly more than previously reported. This could be explained by the study design geared towards actively seeking symptoms of CPSP. The discrepancy between the incidence of CPSP pain as defined by the answer "yes" and PRI 1 is a result of the fact that the latter includes the assessment of numbness.

A limitation of this study is that it was not double-blind, because of ethical considerations, by placing a paravertebral catheter and injecting saline; therefore, it is subject to bias. However, the investigator assessing anxiety and depression, pain, nausea, and pruritus was unaware of the group to which the patient had been assigned. There are a number of confounding variables that may have influenced our results. The heterogeneous nature of surgery (mastectomy and wide local excision) was lessened by the association with axillary node clearance in all cases. The technique used for mastectomy was standardized and performed by the same surgeon. Although a number of patients underwent chemotherapy or radiotherapy, these adjuvant treatments are traditionally commenced between 6 and 8 weeks postoperatively; therefore, they are unlikely to influence the intensity of CPSP at approximately 10 weeks postoperatively. However, the time of telephone interview was carefully chosen between 2 and 3 months (to meet the definition of CPSP), not only to be at the mid-period of the menstrual cycle (when relevant) but also at distance from any recent chemotherapy or radiotherapy to minimize possible mood changes. This may be reflected by the similar HAD scores at this time point compared with postoperative scores. The first blood sample for NOx assay was taken immediately preoperatively, which meant typically 12 hours after the first dose of parecoxib in group N. Although minimizing the variability in parecoxib plasma concentration, it may explain in part the greater variability of NOx production in this group.

Currently, the mechanism of early postoperative pain is poorly understood. The models describing the pathophysiology of sensory changes that occur after injury (largely developed from studies on inflammation in animals), may not translate well to postoperative pain in humans. For example, pain from incisions may be related to ischemia; inflammation may be less critical than originally proposed (13). Much of the pain after breast surgery has historically been attributed to nerve damage (whether from surgery or radiation) (14). Surgical tissue injury also results in spinal sensitization, i.e., metabolic activation and hyperexcitability of spinal nociceptive neurons, expansion of sensory receptive fields and alterations in the processing of innocuous stimuli. These postoperative neuroplastic changes underlie the development of "pathologic" pain, which is characterized by either hyperalgesia, which may be primary or secondary, or allodynia (13).

Kawamata et al. (15) subjected volunteers to a small incision in the volar forearm and mapped the area of hyperalgesia caused by incision. Secondary hyperalgesia (hyperalgesia outside of the injured area) is one measure of enhanced responsiveness of the central nervous system; i.e., central sensitization. The authors noted that the area of flare or redness (possibly a result of axon reflexes) caused by incision was distinct from the area of hyperalgesia. The large area of hyperalgesia did not develop when local anesthetic injection was made before the incision. Although mapping was not done in our study, similar mechanisms may have contributed to less postoperative VAS pain scores in group N, in which a paravertebral block was inserted before incision.

Similarly, pain on movement continued to be less in group N up to postoperative day 5, when COX-2 inhibitors were stopped and none of these patients developed CPSP. Evidence suggests a substantial increase in COX-2 levels in the spinal cord hours after peripheral injury (16). Immediate postoperative COX-2 inhibition may be beneficial by reducing prostanoid production and subsequent intracellular changes in neurons that might contribute to persistent pain (16).

NO plays a role in both the development and maintenance of hyperalgesia (17). Three alternative mechanisms have been proposed to account for NO-induced nociceptor sensitization: i) NO may enhance the release of an algesic substance, i.e., prostaglandin E₂, ii) NO may inhibit the action of an endogenous antinociceptive substance that acts on peripheral nociceptors, or iii) NO might act directly on the nociceptors (8). Furthermore, pharmacologic studies indicate that central sensitization is at least partially mediated by activation of N-methyl-d-aspartate receptors, which could lead to ultimate NO production. Although the link between local and systemic NO production is unclear, a recent study showed that serum NO levels were higher in patients with dysmenorrhea, indicating that there is a role for systemic NO in pain syndromes (9). The overall perioperative profile of NOx after breast surgery in our study is similar to that after other types of surgery (18), with a marked decrease at 12 hours postoperatively. The fact that no further difference between groups could be detected may have been attributable to the small number of patients in each group.

CPSP is a common, important, and under-recognized problem. The most striking postoperative factor predictive of CPSP is the severity of acute pain after thoracic surgery and hernia repair. Our results provide additional evidence of this in patients who underwent breast surgery with axillary node clearance. The association between early postoperative pain and chronic pain does not necessarily imply causality, but it deserves further investigation both to improve understanding and to guide prevention of CPSP.

APPENDIX

Questionnaire used at approximately 10 wk postoperatively (telephone interview).

1. Do you have chronic pain as a result of your breast surgery?
   
2. When did pain begin after your surgery? Immediately/weeks/months
   
3. Where is the location of the pain? a specific location in my breast explain where all over my breast and chest wall in my arm explain where
   
4. Is the pain constant/intermittent?
   If the pain is intermittent, how long does it last? Seconds/min/hrs/days
   
5. Is the pain related to any physical activity such as exercise, movements, clothes? Yes/no
   
6. Are you taking any medication for your pain? List
   
7. Does the medication relieve your pain? Yes/no
   
8. How much does the pain disturb your daily life? Not at all/a little/to some extent/quite a lot/a lot
   
9. Does the pain disturb your sleep? Did you have any chemo- or radiotherapy? When?
   

Footnotes

Accepted for publication June 2, 2006.

REFERENCES

1. Crombie IKK, van Korff M, Linton SJ, et al. Epidemiology of pain: For the IASP Task Force on Epidemiology, 1999.
2. Sharrock NE, Cazan MG, Hargett MJL, et al. Changes in mortality after total hip and knee arthroplasty over a ten year period. Anesth Analg 1995;80:242–8.[Abstract]
3. Macrae WA. Chronic pain after surgery. Br J Anaesth 2001; 87:88–98.[Abstract/Free Full Text]
4. Wallace MS, Wallace AM, Lee J, Dobke MK. Pain after breast surgery: a survey of 282 women. Pain 1996;66:195–205.[Web of Science][Medline]
5. Ivens D, Hoe AL, Podd TJ, et al. Assessment of morbidity from complete axillary dissection. Br J Cancer 1992;66:136–8.[Web of Science][Medline]
6. Katz J, Jackson M, Kavanagh BP, Sandler AN. Acute pain after thoracic surgery predicts long-term post-thoracotomy pain. Clin J Pain 1996;12:50–5.[Web of Science][Medline]
7. Liem MSL, van den Graaf Y, van Steelsel CJ, et al. Comparison of conventional anterior surgery and laparoscopic surgery for inguinal hernia repair. N Engl J Med 1997;336:1541–7.[Abstract/Free Full Text]
8. Anbar M, Gratt BM. Role of nitric oxide in the physiology of pain. J Pain Symptom Manage 1997;14:225–54.[Web of Science][Medline]
9. Sun MF, Huang HC, Lin SC, et al. Evaluation of nitric oxide and homocysteine levels in primary dysmenorrheal women in Taiwan. Life Sci 2005;76:2005–9.[Web of Science][Medline]
10. Zigmmond AS, Snaith RP. The Hospital Anxiety Depression Scale. Acta Psychiatr Scand 1983;67:361–70.[Web of Science][Medline]
11. Melzack R. The McGill Pain Questionnaire: major properties and scoring methods. Pain 1975;1:277–99.[Web of Science][Medline]
12. Perkins FM, Kehlet H. Chronic pain as an outcome of surgery. Anesthesiology 2000;93:1123–33.[Web of Science][Medline]
13. Brennan TJ. Frontiers in translational research. Anesthesiology 2002;97:535–7.[Web of Science][Medline]
14. Tasmuth T, von Smitten K, Hietanen P et al. Pain and other symptoms after different treatment modalities of breast cancer. Ann Oncol 1995;6:453–9.[Abstract/Free Full Text]
15. Kawamata M, Watanabe H, Nishikawa K et al. Different mechanisms of development and maintenance of experimental incision-induced hyperalgesia in human skin. Anesthesiology 2002;97:550–9.[Web of Science][Medline]
16. Samad TA, Sapirstein A, Woolf CJ. Prostanoids and pain: unravelling mechanisms and revealing therapeutic targets. TRENDS in Molecular Med 2002;8:390–6.
17. Salter M, Strijbos PJ, Neale S et al. The nitric oxide-cyclic GMP pathway is required for nociceptive signaling at specific loci within the somatosensory pathway. Neuroscience 1996;73:649–55.[Web of Science][Medline]
18. Iohom G, Szarvas S, Larney V, et al. Perioperative plasma concentrations of stable nitric oxide products are predictive of cognitive dysfunction after laparoscopic cholecystectomy. Anesth Analg 2004;99:1245–52.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Iohom, G. Articles by Shorten, G. D. Search for Related Content PubMed PubMed Citation Articles by Iohom, G. Articles by Shorten, G. D. Related Collections Mechanisms Pain Pharmacology