Year : 2012  |  Volume : 15  |  Issue : 2  |  Page : 122--127

Intrathecal morphine is superior to intravenous PCA in patients undergoing minimally invasive cardiac surgery

Chirojit Mukherjee1, Eva Koch1, Joergen Banusch1, Markus Scholz2, Udo X Kaisers3, Joerg Ender1,  
1 Department of Anesthesiology and Intensive Care Medicine II, Heartcenter, University of Leipzig, Germany
2 Institute of Medical Informatics, Statistics and Epidemiology, University of Leipzig, Germany
3 Department of Anesthesiology and Intensive Care Medicine Medical Faculty, University of Leipzig, Germany

Correspondence Address:
Joerg Ender
Department of Anesthesiology and Intensive Care Medicine II, Heart Center Leipzig, University of Leipzig, Strümpelstrasse 39, 04289 Leipzig


Aim of our study was to evaluate the beneficial effect of low dose intrathecal morphine on postoperative analgesia, over the use of intravenous patient controlled anesthesia (PCA), in patients undergoing fast track anesthesia during minimally invasive cardiac surgical procedures. A randomized controlled trial was undertaken after approval from local ethical committee. Written informed consent was obtained from 61 patients receiving mitral or tricuspid or both surgical valve repair in minimal invasive technique. Patients were assigned randomly to 2 groups. Group 1 received general anesthesia and intravenous patient controlled analgesia (PCA) pump with Piritramide (GA group). Group 2 received a single shot of intrathecal morphine (1.5 μg/kg body weight) prior to the administration of general anesthesia (ITM group). Site of puncture was confined to lumbar (L1-2 or L2-3) intrathecal space. The amount of intravenous piritramide used in post anesthesia care unit (PACU) and the first postoperative day was defined as primary end point. Secondary end points included: time for tracheal extubation, pain and sedation scores in PACU upto third postoperative day. For statistical analysis Mann-Whitney-U Test and Fishers exact test (SPSS) were used. We found that the demand for intravenous opioids in PACU was significantly reduced in ITM group (P <0.001). Pain scores were significantly decreased in ITM group until second postoperative day (P <0.01). There was no time delay for tracheal extubation in ITM group, and sedation scores did not differ in either group. We conclude that low dose single shot intrathecal morphine provides adequate postoperative analgesia, reduces the intravenous opioid consumption during the early postoperative period and does not defer early extubation.

How to cite this article:
Mukherjee C, Koch E, Banusch J, Scholz M, Kaisers UX, Ender J. Intrathecal morphine is superior to intravenous PCA in patients undergoing minimally invasive cardiac surgery.Ann Card Anaesth 2012;15:122-127

How to cite this URL:
Mukherjee C, Koch E, Banusch J, Scholz M, Kaisers UX, Ender J. Intrathecal morphine is superior to intravenous PCA in patients undergoing minimally invasive cardiac surgery. Ann Card Anaesth [serial online] 2012 [cited 2020 May 26 ];15:122-127
Available from:

Full Text


Patients experience severe pain after cardiac surgery and the intensity of pain is maximum on the first postoperative day. [1] The surgical access of anterolateral thoracotomy for minimally invasive cardiac surgery can stimulate severe post operative pain thereby increasing morbidity. The use of lateral thoracotomy involves resection and/or surgical incision of thoracic cage causing pain. However this minimally invasive approach is known to reduce postoperative pulmonary complications. [2]

Pain control is crucial in a fast track concept with short - acting opioids, to facilitate early extubation and a short stay in intermediate care (IMC) unit. Previous studies have demonstrated the positive effects of high doses of intrathecal morphine on analgesia and respiratory function. [3],[4] Jacobson showed that early extubation was possible with a dosage of 6 μg·kg -1 morphine. [5] The minimal effective dose of intrathecal morphine in cardiac surgery still remains unclear.

The aim of our study was to evaluate the effect of 1.5 μg/kg intrathecal morphine, additional to intravenous patient controlled analgesia (PCA) versus PCA alone, within our institutional fast track concept for minimally invasive cardiac surgery [6] on:

Pain score after minimally invasive cardiac surgeryPost-operative opioid consumption, andDuration of extubation

 Material and Methods

After approval from local ethical committee (Ethical Review Board at the Medical Faculty of the University Leipzig, registration number 270-08) we performed a prospective, randomized study. After obtaining patient`s written, informed consent patients scheduled for elective minimally invasive mitral or tricuspid valve repair surgery were included for the study. Exclusion criteria were age < 18 years, pregnancy, known coagulopathies and increased intracranial pressure. Patients having history of previous surgical intervention at spinal T12-L5 level, or having significant spinal deformity and/or neurological deficit were also excluded from the study. Patients were randomized by sealed envelope technique on the day of surgery to receive either general anesthesia with preoperative administration of intrathecal morphine (ITM) or general anesthesia only (GA). Blinding the patients was not possible as this would require administration of an intrathecal placebo which would not have been approved by ethical committee.

Anesthesia and procedure

Oral premedication consisted of 20-40 mg dipotassium-chlorazepate on the evening prior to surgery. Midazolam up to 7.5 mg was given 30 minutes before induction of anesthesia. In the ITM group, intrathecal administration of morphine was performed before induction of general anesthesia with the patient in sitting position. 1,5 μg per kilogram bodyweight of morphine diluted in 10 ml saline was injected into the intrathecal space at L1-2 or L2-3 interspace, using an atraumatic 25 G needle (Pajunk® , Germany). Control group did not receive any intrathecal morphine. In an event of a traumatic or bloody tap, surgery was postponed.

After pre- oxygenation, anesthetic induction was performed with bolus injection of propofol (1-2 mg/kg), fentanyl (0.5-1 μg/kg), and rocuronium (0.6 mg/kg). Maintenance of anesthesia was achieved with continuous infusion of remifentanil (0.2 μg/kg/min). Sevoflurane 0.8-1.1% minimum alveolar concentration (MAC) was used pre- and post cardiopulmonary bypass whereas during cardiopulmonary bypass, a continuous infusion of propofol (3 mg/kg/h) was administered.

A recruitment maneuver was carried out prior to weaning from cardiopulmonary bypass to prevent atelectasis. An external warming system with an underbody blanket (Bairhugger® ; Arizant Healthcare, Eden Prairie, MN) was used after weaning from cardiopulmonary bypass to ensure a minimum core temperature of 36°C. For early postoperative analgesia, 1 g intravenous paracetamol was administered to every patient before skin closure. Criteria for transfer to the post anesthesia care unit (PACU) were

Minimal or no inotropic supportCore temperature of ≥36°CNo active bleeding.

Patients were transferred to the PACU with endotracheal tube in place and a reduced infusion rate of remifentanil (0,1μg/kg/min). Extubation was performed after assessment by the attending physician. All patients underwent an elective non - invasive ventilation for 60 minutes post extubation (Elisee 350® , Saime, France).

After arrival in PACU, patients in control group received a bolus of 0.1g/kg of piritramide, to ease the change from the short acting opioid remifentanil to a longer acting opioid piritramide for postoperative pain therapy. This management is part of the institutional fast track concept. [6] Furthermore from our previous experience of more than 200 patients we found that routine administration of bolus of piritramide on arrival at PACU significantly delayed extubation in patients who received intrathecal morphine pre-operatively. Therefore blinding of the patient was not possible. Sedation score in PACU, IMC unit and on transfer to wards was assessed with our institutional scoring system as follows:

Awake and cooperativeImmediate response to verbal commandsLethargic response and responsive only to loud verbal commandsResponse only to painful stimuliUnarousable.

For further postoperative pain control in both groups, an intravenous PCA pump containing piritramide [7] at a concentration of 1mg/ml was used. PCA pump was programmed to administer 2 ml bolus of piritramide on demand, without a continuous basal infusion, and a lockout period of 10 minutes between each bolus administration. The attending physician was unaware of the patients to whom intrathecal morphine was administered. All patients were further transferred to IMC for an overnight monitoring prior to discharge to surgical-floor the following day. In IMC unit facilities for continuous invasive hemodynamic monitoring was possible but without the facility of invasive or non invasive ventilation. For transfer to IMC unit, patients need to be awake, oriented and in a stable hemodynamic condition without any inotropic support. A sufficient gas exchange (p a CO 2 < 50 mmHg, p a O 2 > 70 mmHg) under insufflation of 3-5 l/min O 2 was necessary for admission to IMC unit.

Criteria for transfer to wards included the presence of a stable rhythm and independent mobilization. Criteria for prolongation of IMC unit stay were: insufficient diuresis, intravenous amiodarone therapy, dependence on external pacemaker or symptomatic transitory psychotic syndrome. Intravenous PCA was continued in the ward and was terminated by the pain service of the anaesthetic department, in accordance with visual analogue scale (VAS) for pain.

Data assessment

Data were assessed on the day of operation and following three postoperative days. All parameters for PCA were documented till third postoperative day.

Extubation time was defined as the time from arrival in the PACU to tracheal extubation. Blood gas analysis were registered every hour in the first 4 hours after extubation as well as every 8 hours on IMC unit. Postoperative pain intensity was evaluated at rest, using a visual analogue scale (VAS) from 0 to 10. Sedation scores were assessed simultaneously on a numeric scale reaching from 1 (awake) to 5 (unarousable). The amount of intravenous opioids administered in PACU and IMC Unit was documented. The consumption of PCA piritramide was calculated according to the number of boluses utilized till the 3 rd postoperative day. The appearance of possible side effects associated with intrathecal morphine like nausea, vomiting, pruritus, headache and urinary retention, was recorded.


The primary endpoint of this study was the amount of intravenous opioids administered on the day of surgery and first postoperative day. Sample size of the study was calculated based on retrospective data regarding this endpoint. Based on this data, we calculated that a sample size of N=28 is sufficient for a 90% power at 5% significance level. For power calculation PASS 2008, Version 08.0.5 was used. We assumed a drop-out rate of 10% of patients who are not eligible for fast-track after surgery. Hence, a total of N=31 patients per group were randomized to the two treatment arms.

Time to tracheal extubation, pain scores, and sedation scores were analyzed as secondary endpoints.

For continuous variables we used the Mann- Whitney- U- test and for categorical variables we used Fishers exact test in order to compare the two treatment groups. Analysis was performed using SPSS (SPSS version19, Chicago, IL, USA).

A P value < 0.05 was considered to be significant.


Demographic data

Between January 2009 and November 2011, sixty-one out of the 62 randomized patients scheduled for minimally invasive cardiac surgery were included in the study. Of the randomized 62 patients, one patient developed postoperative hemorrhage and was excluded from the study. Demographic data between the groups showed no significant differences [Table 1].{Table 1}

Operative data

All lumbar punctures were performed successfully. There was no significant difference in surgery time, duration of cardiopulmonary bypass or aortic cross clamp time [Table 2]. Postoperative hemorrhage occurred in one patient and he was excluded from the study.{Table 2}

Postoperative data

The median time to extubation and the median CO 2 partial pressure showed no significant differences between the groups. There was one case of reintubation in ITM group. Sedation scores in PACU did not differ among the two groups [Table 3].{Table 3}

Length of stay in PACU, IMC unit and hospital as well as time to mobilization were similar in both groups [Table 3].

Analgesia and side effects

The amount of piritramide administered in PACU was significantly reduced in ITM group (P <0.001). Piritramide requirements per day in patients with intrathecal morphine administration were lower as compared to GA group until second postoperative day, although statistically not significant [Figure 1].{Figure 1}

There was a significant decrease of pain intensity in ITM group versus GA group until second postoperative day [Figure 2].{Figure 2}

Opioid induced nausea and vomiting was similar in both groups [Table 3].


The study demonstrated that

Intrathecal morphine administration at a minimal dosage of 1.5 μg per kg can reduce the additional opioid requirements during the immediate post-operative phase in comparison to control group.Intrathecal morphine administration can provide adequate analgesia for post operative pain management when compared to control group.The time for extubation is not delayed with the administration of intrathecal morphine in comparison to the control group.

Methods and procedure

Our study size included 62 patients. We believe our prospective randomized study has consistently demonstrated effective pain relief even with low dose intrathecal morphine when compared to results using high dose opioids. The patient population although using higher dosage of opioids had similar number of patients when compared to our study, but very few prospective randomized studies has been published to this effect. [8],[9] Many substances and techniques for neuro-axial analgesia have been tried with different impact on postoperative patient comfort and morbidity. Compared with other opioids like sufentanil and fentanyl, morphine is less hydrophobic, therefore, the onset of an analgesic effect is delayed (45-75 minutes), with long lasting effect compared to fentanyl or sufentanil and its distribution in the cerebrospinal fluid (CSF) is also widespread. [10] Using an injecting volume of 10cc, the cephalic spread in the subarachnoid space can be enhanced, owing to a larger displaced volume allowing easier access to nerve roots. Moreover the site of puncture at L1-L2 achieves a wider distribution of morphine. The CSF volume above L2 level is lower as the spinal cord occupies a major portion of the subarachnoid space. The injected solution therefore ascends in cephalic direction [11] and permits a very minimal dosage of 1.5 μg/kg in achieving adequate analgesia. The risk of neurological complications associated with a spinal puncture at lumbar level L1-L2 is well known. Regarding the results of a recent magnetic resonance imaging study, the conus medullaris terminus (CMT) in non spinal disease patients is located at the height of the middle and lower third of the first lumbar body. [12] Since we were performing the puncture below that level, we considered it to be safe. According to recent analysis, the estimated risk of haematoma formation lies in the range of 1:3600 and 1: 220.000 with 95% confidence. [12] We did not experience any traumatic puncture or postoperative haematoma in our study population. However our study population cannot negate this possibility due to small number of patients involved in the study.

Postoperative analgesia

Analysis of pain scores and iv. opioid demand in PACU revealed superiority of intrathecal morphine during the early postoperative hours. This is in agreement with previous studies. [3],[5] A plausible explanation for significant decrease of pain intensity until second postoperative day in our study may be a pre-emptive analgesic effect induced by intrathecal morphine. By definition pre-emptive analgesia is supposed to delay the development of central sensitization after incision or inflammatory injury, with surgical onset and persists over the early postoperative period. [13] The clinical relevance of this phenomenon remains controversial; however our findings reflect the outcomes of a meta-analysis. In the study of Ong and colleagues epidural anesthesia was performed to obtain a significant pre-emptive effect by attenuating postoperative pain scores, reducing the demand for additional analgesic and delaying early rescue analgesia. [14]

The opioid consumption in ITM group on third postoperative day, does not reflect worsening of pain, as pain scores were on a decrease scale. Actually, patients (4 in ITM and one in GA group) reported, they had used PCA at night for prophylactic use in anticipation of pain and to facilitate sleep.

Extubation times and side effects

There was no difference in median extubation times between the groups indicating that the use of intrathecal morphine does not interfere with our concept of early extubation if no routine bolus of piritramide is given after arrival on PACU. [6]

Although median CO 2 partial pressures provide no evidence of respiratory depressive effect of ITM, one case of reintubation occurred in ITM group. The patient received in total 115 μg intrathecal morphine. After uncomplicated early extubation (extubation time = 60 min), the patient has to be re-intubated immediately after administration of additional 4 mg of piritramide. This bolus of piritramide led to immediate apnoea. The patient could be extubated 180 min after reintubation. This underlines our previous experience that patients who received intrathecal morphine preoperatively do not need routinely an additional bolus of piritramide without expressing pain. Six cases of postoperative vomiting occurred in ITM group and 4 in GA group which was not significant. It is unlikely that this is a side effect associated exclusively with intrathecal morphine administration. Since vomiting occurred in both groups and not only on the day of operation, it may also be triggered by piritramide, which has an equal emetic potential as morphine. [15] Two patients reported explicitly about vomiting after piritramide administration via PCA pump. As all patients received intraoperative transesophageal examination the possibility of TEE induced postoperative vomiting cannot be excluded.

Study limitations

Study design did not permit placebo puncture, therefore a bias towards ITM group cannot be excluded. However, we did not consider a puncture with saline to be ethically justifiable in patients who will be heparinised for cardiopulmonary bypass and would not have been approved by the local ethics committee. To prove the safety of intrathecal puncture at this relatively high level the number of patients is far too small.


Our study demonstrated that low-dose intrathecal morphine administration provides adequate analgesia during the early postoperative period. The dosage requirements of intravenous PCA drug is relatively lower in this group compared to the control group. Preoperative administration of intrathecal morphine does not prevent early extubation and therefore it is justified to consider it as an alternative therapeutic option compared to PCA alone in minimally invasive cardiac surgery.

A larger study population is required to reliably conclude and distinguish between the side effects of opiod related complications and the procedure itself.


M.S. is funded by LIFE - Leipzig Research Center for Civilization Diseases, Universitδt Leipzig. LIFE is funded by means of the European Union, by the European Regional Development Fund (ERDF) and by means of the Free State of Saxony within the framework of the excellence initiative.


1Chaney MA. Intrathecal and epidural anesthesia and analgesia for cardiac surgery. Anesth Analg 2006;102:45-64.
2Weissman C. Pulmonary complications after cardiac surgery. Semin Cardiothorac Vasc Anesth 2004;8:185-211.
3Bettex DA, Schmidlin D, Chassot PG, Schmid ER. Intrathecal sufentanil-morphine shortens the duration of intubation and improves analgesia in fast-track cardiac surgery. Can J Anesth 2002;49:711-7.
4dos Santos LM, Santos VC, Santos SR, Malbouisson LM, Carmona MJ. Intrathecal morphine plus general anesthesia in cardiac surgery: effects on pulmonary function, postoperative analgesia, and plasma morphine concentration. Clinics (Sao Paulo). 2009;64;279-85
5Jacobsohn E, Lee TW, Amadeo RJ, Syslak PH, Debrouwere RG, Bell D, et al. Low-dose intrathecal morphine does not delay early extubation after cardiac surgery. Can J Anesth 2005;52:848-57.
6Ender J, Borger MA, Scholz M, Funkat AK, Anwar N, Sommer M, et al. Cardiac surgery fast-track treatment in a postanesthetic care unit: six-month results of the Leipzig fast-track concept. Anesthesiology 2008;109:61-6.
7Kumar N, Rowbotham DJ: Piritramide. Br J Anaesth 1999;82:3-5
8Mehta Y, Kulkarni V, Juneja R, Sharma KK, Mishra Y, Raizada A, et al. Spinal (subarachnoids) morphine for off-pump coronary artery bypass surgery. Heart Surg Forum 2004;7:E205-10.
9Chaney MA, Smith KR, Barclay JC, Slogoff S. Large-dose intrathecal morphine for coronary artery bypass grafting. Anesth Analg 1996;83:215-22.
10Rathmell JP, Lair TR, Nauman B. The role of intrathecal drugs in the treatment of acute pain. Anesth Analg 2005;101(5 Suppl):S30-43.
11Greene NM. Distribution of local anesthetic solutions within the subarachnoid space. Anesth Analg 1985;64:715-30.
12Lin N, Bebawy JF, Hua L, Wang BG. Is spinal anesthesia at L2-L3 interspace safe in disorders of the vertebral column? A magnetic resonance imaging study. Br J Anaesth 2010;105:857-62.
13Kissin I. Preemptive analgesia. Anesthesiology 2000;93:1138-43.
14Ong CK, Lirk P, Seymour RA, Jenkins BJ. The efficacy of preemptive analgesia for acute postoperative pain management: A meta analysis. Anesth Analg 2005;100:757-73
15Breitfeld C, Peters J, Vockel T, Lorenz C, Eikermann M. Emetic effects of morphine and piritramide. Br J Anaesth 2003;91:218-23.