A Randomized, Double-Blinded Comparison of Intrathecal Morphine, Sufentanil and their Combination versus IV Morphine Patient-Controlled Analgesia for Postthoracotomy Pain

Abstract

We compared the analgesic effect of lumbar intrathecal (IT) 0.5 mg morphine (Group M, n = 10), 50 μg sufentanil (Group S, n = 10), and their combination (Group S-M, n = 10) given before general anesthesia and patient-controlled analgesia with IV morphine (Group C, n = 19) in a randomized, double-blinded study performed in patients undergoing thoracotomy. Pain visual analog scale (VAS) and morphine consumption were assessed for 24 h. In Group S-M the number of patients initially titrated with IV morphine was less than in group C (30 vs 84%, P < 0.05). Morphine requirement was higher in Group C (71 ± 30 mg) than in Groups S (46 ± 34 mg, P < 0.05), M (38 ± 31 mg, P < 0.05) and S-M (23 ± 16 mg, P < 0.01). VAS scores were significantly decreased during the first 0–11 postoperative h at rest and during the first 0–8 postoperative h on coughing in Groups M and S-M rather than in Group C. The incidence of side effects was infrequent except for urinary retention. Preoperative IT morphine or combined sufentanil and morphine could be given as a booster to achieve rapidly effective analgesia in the immediate postoperative period.

IMPLICATIONS 

As compared with IV patient-controlled analgesia, intrathecal morphine or combined sufentanil and morphine provided superior postoperative pain relief both at rest (11 h) and on coughing (8 h) than did IV patient-controlled analgesia morphine alone. IV morphine requirement was decreased during the first postoperative day after posterolateral thoracotomy.

Lumbar intrathecal (IT) opioids have been used as an adjunct to postthoracotomy analgesia ^(1–5), but the available literature ⁽⁶⁾ has not clearly demonstrated the clinical usefulness of the technique. No study has compared IT morphine to IV patient-controlled analgesia (IVPCA) with morphine, particularly in the immediate postoperative period.

The observation of delayed analgesia at the thoracic dermatomes after intrathecal morphine ⁽¹⁾ is consistent with the peak effect of IT morphine, which occurs some 6–7 h after injection ⁽⁷⁾. Sufentanil IT provides intense and almost instantaneous analgesia lasting a mean of 5 to 7 h ⁽⁸⁾, sufficient to cover the intraoperative and early postoperative periods, whereas IT morphine provides prolonged postoperative analgesia lasting approximately 24 h, but with delayed onset.

We therefore postulated that combining IT sufentanil and IT morphine would provide adequate analgesia for the first 24 h. The aim of this prospective randomized double-blinded study was to compare the analgesia provided by IT sufentanil, IT morphine, and their combination, and IVPCA with morphine during the first 24 h after posterolateral thoracotomy.

Materials and Methods

The Ethical Committee of our University approved this prospective randomized double-blinded study, in which all physicians and nursing staff in charge of the patients during and after surgery were blinded as to treatment.

Written informed consent was obtained from 50 patients scheduled for primary elective lung surgery via a posterolateral midthoracic incision without costectomy. Patients who were under 18 or over 70 yr of age, weighed <50 kg or >100 kg, had an ASA physical status more than III, had an active infectious process, a neurological disorder, were receiving opioid therapy for chronic pain, had abnormal coagulation tests, or were to undergo a surgical procedure starting after 2 pm were excluded. After enrollment, the patients were randomly allocated to one of the following four groups: Group C = control group with a local anesthesia of the skin over the lumbar spine, Group S = IT sufentanil, Group M = IT morphine and Group S-M = combined IT sufentanil and morphine. A five-group randomization table was used. Two randomly selected groups were considered the control group, and each of the three remaining groups as a treatment group. The IT study solutions consisted of either 50 μg sufentanil ( Sufenta®; Janssen-Cilag, Berchem, Belgium) plus 5 mL 0.9% saline (Group S); or 0.5 mg preservative-free morphine plus 5 mL 0.9% saline (Group M); or 50 μg sufentanil plus 0.5 mg preservative-free morphine plus 4 mL 0.9% saline (Group S-M). Each study solution was of 6 mL.

On the day before surgery, patients received instructions on how to use a patient-controlled analgesia (PCA) device ( Lifecare 4200® PCA infuser; Abbott Laboratories, North Chicago, IL), and a pain visual analog scale (VAS) that consisted of an unmarked 100-mm line, with 0-mm representing no pain and 100-mm representing the worst pain imaginable.

All patients received oral hydroxyzine (1.5 mg/kg) for preoperative sedation 1 h before being transferred to the operating room and had their usual medications continued through the morning of surgery. Once IV access had been achieved, all patients were placed in the sitting position. The skin surrounding the lumbar area was prepared with a Betadine solution and draped. Local anesthesia of the skin was performed with 2–3 mL of 1% lidocaine at the L3-L4 or L4-L5 interspace. For the IT opioid groups, a 24-gauge spinal needle with pencil point ( Pencan®; Braun, Melsungen, Germany) was used. After return of clear, free-flowing cerebrospinal fluid (CSF), the IT solution was injected into the subarachnoid space over 5–10 s. Verification of the ability to aspirate CSF was made after the solution was injected, and the needle was then removed.

Immediately after the IT injection, general anesthesia was induced in all patients with thiopental (5–7 mg/kg) and fentanyl (1–1.5 μg/kg). Rocuronium (0.6 mg/kg) was administered to facilitate endobronchial intubation. Anesthesia was maintained with nitrous oxide (inspired concentration adjusted to oxygen saturation) and isoflurane (up to MAC 1.2 monitored with Capnomac®, Datex, Helsinki, Finland). Rocuronium increments were used to facilitate mechanical ventilation of the lungs. Whenever the blood pressure or heart rate increased by more than 30%, so that depth of anesthesia was judged inadequate, hemodynamic control was ensured with incremental doses of IV fentanyl (50–100 μg). Two surgeons with similar techniques performed all the operations. One (in case of pneumonectomy) or two lateral chest tubes (lobectomy) were placed in every patient at the end of the surgical procedure.

Isoflurane administration was stopped at the beginning of parietal closure. At the end of the procedure, the patients were awakened and their tracheas extubated if they met standard extubation criteria (regular breathing, end-tidal carbon dioxide <45 mm Hg and Spo₂ more than 95%). The patients were transferred to the recovery room for close monitoring over the next 24 h (heart rate, blood pressure, oxyhemoglobin saturation, respiratory rate by thoracic impedance or nasal capnography, and level of consciousness). All patients received oxygen via a face mask to maintain Spo₂ more than 90% throughout the study period.

When sufficiently awake for pain assessment, patients were asked to score pain on the VAS. This evaluation defined T0. A score more than 30 mm indicated pain requiring an initial titration with IV morphine administered by a nurse and consisting in 5 mg of morphine every 5 min until the VAS pain score at rest was ≤30 mm. IVPCA was then started with a PCA device programmed to deliver morphine in bolus doses of 1.5 mg, with a lockout duration of 3 min and an unrestricted total dose. Patients whose VAS pain score at T0 was ≤30 mm, the limit that defined effective analgesia, were started directly on IVPCA. Patients received no sedatives or opioids other than IVPCA morphine. IVPCA morphine consumption was recorded in all patients every hour from T0 to T12 h and then every 2 h until T24 h. The number of patients who received an initial titration of IV morphine and the total amount of morphine given were recorded.

Pain at rest was scored on the VAS every hour from T0 to T12 h and every 2 h for T12 to T24 h. Pain on coughing was scored every 4 h on the VAS. The duration of effective analgesia was defined as the sum of each period during which VAS pain score was ≤30 mm and was only calculated for pain at rest, not for pain on coughing because VAS scores were only measured every 4 h.

The evaluator rated the degree of patient sedation on a five-point scale (1 = wide awake, 2 = drowsy or dozing intermittently, 3 = mostly sleeping but easily awakened, 4 = asleep, difficulty responding to verbal commands, 5 = awakened only by shaking) every 4 h. In case of excessive sedation (sedation score equal or greater than 4) combined with a respiratory rate ≤8/min, IV naloxone was administered. Episodes of pruritus, nausea, vomiting and urinary retention, requiring bladder catheterization, were recorded over the 24 h study period.

Continuous data (age, height, weight, and duration of satisfactory analgesia) were compared by analysis of variance using the Kruskal-Wallis test. Comparison of serial measurements (VAS for pain relief and sedation score) was performed with repeated-measures analysis of variance. Ranked data were analyzed with the Kruskal-Wallis and Mann-Whitney U-tests when appropriate. Categorical data were examined by Fisher’s exact test. Probability values under 0.05 were considered significant. Data are expressed as mean ± sd.

Results

Fifty patients were included, 20 in Group C and 10 each in Groups S, M, and S-M. One patient in Group C was reoperated 2 h after his arrival in the recovery room because of excessive bleeding and was excluded from the study. Forty-nine patients were therefore prospectively studied for 24 h. The four groups were similar with respect to demographic variables (Table 1). Types of surgery, procedure duration, and mean intraoperative fentanyl doses are given in Table 1. The intraoperative fentanyl requirements were significantly less in Group S-M when compared with Group C (135 ± 58 μg vs 266 ± 101 μg, P < 0.05).

Table 1: Demographic and Intraoperative Data

The trachea could not be extubated in one patient in each group at the end of the procedure, but was extubated 10 to 30 min after arrival in the recovery room. These patients were retained in the study. The percentage of patients who needed initial titration with IV morphine was significantly smaller (P < 0.05) in Group S-M (30%) in comparison with Group C (84%), but was not different in Groups S (60%) and M (50%) in comparison with Group C. The titrated morphine dose was significantly smaller in each of the three IT opioid groups than in Group C, with no difference among the IT opioid groups (17 ± 10 mg in Group C, 8 ± 12 in Group S, 5 ± 6 mg in Group M, and 5 ± 9 mg in Group S-M;P < 0.05 Groups S, M, and S-M versus Group C). The total 24-h IV morphine requirements, including the titrated dose, were also significantly lower in each of the 3 IT opioid groups than in Group C, with no difference among them (71 ± 30 mg in Group C, 46 ± 34 in Group S, 38 ± 31 mg in Group M, and 23 ± 16 mg in Group S-M;P < 0.05 Groups S, M, and S-M versus Group C (Fig. 1).

Figure 1: Cumulative IV morphine requirements during the 24 h postoperatively. Group C = local anesthesia of the skin, Group S = IT sufentanil, Group M = IT morphine, and Group S-M = combined IT sufentanil and morphine. Data are expressed as mean ± sem.* P < 0.05, ** P < 0.01.

At rest, the VAS pain scores were significantly decreased during the first 3 h in Group S (P < 0.05) and over the first 11 h in Groups M and S-M (P < 0.01) in comparison with Group C. At T12 h, pain relief was no longer statistically different in the four groups (Fig. 2). The duration of effective analgesia (a VAS pain score under 30 mm) was significantly longer in Groups S, M, and S-M in comparison with Group C (20.4 ± 2.4 h, 21.9 ± 2.8 h, 22.7 ± 1.6 h, and 16.3 ± 5 h respectively;P < 0.05 Group S versus Group C, P < 0.001 Groups M and S-M versus Group C). In addition, the duration of effective analgesia was significantly longer in Group S-M when compared with Group S (P = 0.02).

Figure 2: Change in pain at rest (mean ± sem), as measured by a 10-cm visual analog scale (VAS) during the first 24 h. Group C = local anesthesia of the skin, Group S = IT sufentanil, Group M = IT morphine, and Group S-M = combined IT sufentanil and morphine. * P < 0.05, ** P < 0.01.

The VAS pain scores on coughing were significantly lower from T0 to T4 h in Group S and from T0 to T8 h in Groups M and S-M, in comparison with Group C (P < 0.05) (Fig. 3).

Figure 3: Change in pain on coughing (mean ± sem), as measured by a 10-cm visual analog scale (VAS) during the first 24 h. Group C = local anesthesia of the skin, Group S = IT sufentanil, Group M = IT morphine, and Group S-M = combined IT sufentanil and morphine. * P < 0.05.

The degree of sedation in the four groups was similar over the 24-h study period. There were severely sedated patients (scale 5 = awakened only by shaking) in all the groups, especially during 8–20 h, which corresponds in nearly all cases to the first postoperative night (Fig. 4). No case of respiratory depression was observed (Table 2).

Figure 4: Proportion of patients presenting with each degree of sedation from the time of arrival in the recovery room (T0, 0 h) to the end of the study period (T24, 24 h) in 4-h increments (100% stacked chart). The differences are not statistically significant.

Table 2: Side Effects

The incidence of urinary retention, requiring bladder catheterization, was frequent in all the groups. Furthermore, a statistically significant difference was found between Group C and all three IT opioid groups together (pooled data) (42 vs 75%, P = 0.03) (Table 2). The incidence of nausea and pruritus was infrequent (Table 2). No patient experienced side effects such as postdural puncture headache resulting from lumbar puncture.

Discussion

Our study showed that IT morphine, sufentanil, and their combination provided more effective analgesia than IVPCA morphine during the first postoperative day after posterolateral thoracotomy. IT morphine alone or combined with IT sufentanil provided similar and adequate pain relief at rest and on coughing, with similar titrated and total IV morphine doses. IT sufentanil had a limited analgesic effect over the 3 hours.

IVPCA with morphine had several limitations in our posterolateral thoracotomized patients. Satisfactory analgesia at rest (pain score less than 30 mm on a 0–100 mm scale) occurred only after 6 hours, despite initial titration with large amounts of IV morphine. Moreover, effective analgesia (i.e., a VAS pain score under 30 mm) lasted only 16.3 ± 5.0 hours. Furthermore, the mean pain scores on coughing were greater than 30 mm over the entire study period.

Indications for postoperative analgesia with IT morphine, which is a simple and reliable technique, are very large ⁽⁹⁾, and several studies have reported the value of IT morphine in postthoracotomy pain management ^(1–5,9). However, these studies were uncontrolled and documented IT morphine efficacy using either the time until the first request for an analgesic ⁽²⁾ or the total 24-hour meperidine requirement and the pain score ⁽⁵⁾, but never the IVPCA morphine requirement, which has now become one of the methods used to compare analgesic techniques. Our study demonstrated that preoperative IT injection of 0.5 mg of morphine supplemented by IVPCA morphine provided 21.9 ± 2.8 hours of effective analgesia. However, half of our patients needed an initial titration of IV morphine and pain relief at rest and on coughing, which was significantly better only during the first 11 and 8 hours postoperatively, respectively, compared with IVPCA alone.

Single-bolus IT injection of a lipophilic opioid provides rapid pain relief but its use for postoperative analgesia is limited by a relatively short duration of action ⁽¹⁰⁾. Few studies have investigated the value of IT sufentanil for postthoracotomy analgesia. Hansdottir et al. ⁽³⁾ reported that immediate preoperative IT administration of 15 μg sufentanil produces short duration analgesia, lasting 3 hours after the end of anesthesia. In our study, although a larger dose of sufentanil was administered (50 μg), more than half of the patients needed initial titration with IV morphine. Furthermore, we found that in comparison with IVPCA alone, pain relief at rest and on coughing was significantly better only during the first 4 and 3 hours, respectively.

Combined IT morphine and sufentanil has been proposed in an open study as part of a fast-track technique in selected cardiac surgical patients ⁽¹¹⁾. This technique makes it possible to obtain a painless awakening, a major goal of any analgesic technique, in 70% of our patients. It also decreases intraoperative fentanyl requirements. However, this advantage is only marginal because short-lasting opioids are now available. Furthermore, IT combination did not differ from IT morphine alone; there were similar additional IVPCA morphine requirements and identical analgesic efficacy both at rest, with a mean duration of effective analgesia of 22.7 ± 1.6 hours.

The rationale for combining an IT low lipophilic opioid (morphine) and a highly lipophilic opioid (sufentanil) follows from their pharmacokinetic differences. After IT administration, sufentanil concentrations in the CSF decrease more quickly than do morphine concentrations: mean residence time in the CSF, i.e., the time required to eliminate 63.2% of the drug, is approximately 2.3 hours after 0.05 mg/kg morphine ⁽¹²⁾ and 0.9 hours after injection of 15 μg sufentanil ⁽³⁾. Plasma concentration profiles are also different: small concentrations less than 1 ng/mL are observed after 0.25 or 0.50 mg IT morphine ⁽¹⁾, sufentanil peak plasma concentrations of 0.15 ± 0.02 ng/mL occurring 39 ± 10 min after 15 μg IT sufentanil ⁽³⁾.

Surprisingly, nausea, vomiting, and pruritus were uncommon in our patients, whereas others have reported these complications to occur in approximately half of patients ^(13–14) and in 20%–100% of healthy volunteers after IT morphine ⁽⁷⁾. However, the frequent incidence of urinary retention observed in all four study groups can be explained by the presence of many related factors: large percentage of men ^(15,16) whose mean age was compatible with the presence of benign prostatic hypertrophy, and extensive use of morphine ⁽¹⁶⁾. Furthermore, the incidence of urinary retention was more frequent in the IT opioid groups than in the IVPCA group, as reported previously ⁽¹⁷⁾. IV naloxone to restore voiding capacity ⁽¹⁸⁾ could have been given before bladder catheterization was performed, but we decided against it because it would have interfered with IVPCA morphine evaluation.

The most serious side effect of IT opioid is respiratory depression. No episode of oxygen desaturation (Spo₂ less than 90%) was observed in our patients, but all the patients received oxygen in the postoperative period. Respiratory depression did not occur, although the combination of opioids could have enhanced the risk of immediate and late respiratory depression. Despite the large dose of sufentanil, early respiratory depression, which can result from its rostral diffusion and vascular resorption ⁽¹⁹⁾, could not be observed in our series because recovery always occurred more than 2 h after IT injection. This 2-h period is long enough to allow the plasma sufentanil concentration to decrease ⁽³⁾. Late respiratory depression, which is a result of the rostral redistribution to brainstem respiratory centers, is a complication of IT morphine ⁽⁷⁾. The lack of respiratory depression must be considered within the context of the small number of treated patients, because very large series have reported figures as small as 0.36%^(20,21) or as large as 3%⁽⁹⁾. In the latter study, by Gwirtz et al. ⁽⁹⁾, no patient died or required emergent intubation. Like these authors, our policy is to keep major thoracic surgical patients who have received IT morphine under continuous monitoring (postanesthesia care unit or intensive care unit). Clinical monitoring, respiratory rate and sedation level measurements are essential; pulse oxymetry and nasal capnography, for example, are supplementary devices ⁽²²⁾.

Our study did not include a true placebo group (an IT injection of physiological saline solution) because our Ethical Committee had rejected the idea on the grounds that dural puncture is considered an invasive procedure. Nevertheless, the study was double blinded. Because only one dose of morphine and one dose of sufentanil were used, we cannot distinguish between an additive and a supraadditive effect.

In conclusion, preoperative IT morphine or combined sufentanil and morphine provided superior postoperative pain relief both at rest (11 hours) and on coughing (8 hours), than did IVPCA morphine alone. IT opioids could therefore be given as an initial booster to achieve rapidly effective analgesia in the immediate postoperative period, and either combined or followed by IVPCA that would provide analgesia for the following days. The drawbacks of the technique include postoperative urinary retention requiring bladder catheterization. In addition, careful monitoring for at least 24 hours is mandatory.