August Bier is recognized as the father of intrathecal anaesthesia and, in 1898, he and his assistant performed intrathecal anaesthesia on each other with cocaine at the Royal Chirurgical Clinic, in Kiel. The earliest documented use of intrathecal opioids was by Racoviceanu-Pitesti  of Romania in 1900 and was presented at the Congress of International Surgery in Paris. A combination of morphine and cocaine was used. Intrathecal opioid administration was introduced into European clinical practice in the late 1970s, but UK institutions took several more years to begin using the technique on a regular basis. Concerns over adverse side-effects (mainly respiratory depression) have reduced as lower doses have been employed. The popularity of intrathecal opioid administration has grown and now intrathecal opioids are employed regularly in most anaesthetic departments. Many hospitals have developed local policies, guidelines or protocols for their use. However, there is no consensus of opinion on which opioids are used, the correct dose to maximize pain relief but minimize complications, and where the patients should be nursed postoperatively, irrespective of the surgery performed. The aim of this survey was to identify the current practices of anaesthetic departments within the UK.
The survey was based on a questionnaire sent to all clinical directors of anaesthetic departments in the UK. In all, 270 questionnaires were sent. Each questionnaire was numbered but anonymized. The audit office at the Queen's Medical Centre University Hospital NHS Trust, Nottingham, collected the results. The questionnaire examined: (1) the use of local guidelines or protocols; (2) the opioids used and their doses; (3) the types of surgery performed when intrathecal opioids are used; (4) the destination of the patients postoperatively; (5) the information given to the nursing staff; and (6) any cases of serious complications. A reminder was mailed to non-responders. Data are presented as numbers or percentages of respondents.
In all, 199 replies (73.7%) from the 270 questionnaires were received. One of the returns was not completed at all and is not included in the following statistics. Of the respondents, 175 (88.4%) used intrathecal opioids as part of their department's standard practice. Of those who used intrathecal opioids, 107 (61.1%) had local guidelines/protocols in place; the other 68 (38.9%) did not. The following data are displayed as totals (percentage of total intrathecal opioid users). There were responses from six paediatric hospitals, whose results are considered separately.
Choice of opioids
Table 1 describes the pattern of intrathecal opioid administration throughout the UK. The highest quoted dose of diamorphine was 4 mg by one department.
Of those who responded, 60.8% of departments were using intrathecal opioids in their obstetric unit, 57.6% in orthopaedics, 38.9% in general surgery, 30.8% in gynaecology, 25.3% in vascular surgery, 2.5% in spines, 1% in paediatrics and 9.6% in other departments including urology and thoracic surgery.
Of the responding departments, 55% of those with guidelines or protocols in place, and 70.6% of those without, returned patients to a normal surgical ward. Departments with guidelines or protocols returned a further 14.2% of patients to a critical care area as opposed to 7.4% of departments without guidelines or protocols. The critical care areas included intensive care (ICU), high-dependency (HDU), step-down (SDU) and extended recovery units. The remainder were allocated variably or the respondents gave no answer.
Post-anaesthetic management instructions (Table 2)
Departments with guidelines or protocols in place were likely to have a protocol for postoperative management for the nursing staff to follow, although over a third simply provided written instructions. The two positives for a protocol in the non-guidelines or protocols section had a protocol for obstetric patients but not other specialities. Written guidance usually consisted of anaesthetists writing instructions for the nursing staff on each individual anaesthetic chart. Four of the departments with guidelines or protocols in place did not leave any individualized post-anaesthetic instructions for the nursing staff.
Guidance for further opioid use in the following 24 h was given in 93 (87.7%) departments in the guidelines group and 16 (23.5%) in the non-guideline group. Delayed opioid prescription in 50 (28.7%) departments was the most common method of prescribing, followed by normal opioids with care in 34 (19.5%) with a further 22 (12.6%) departments using a patient-controlled analgesia (PCA) system. A combination of two or all three of the alternatives was used by 17 (9.8%) departments, a further 6 (3.5%) did not answer the question and 43 (25.9%) departments did not have any guidance for future opioid prescribing.
Serious complications or adverse events
There were 22 adverse events; these were the most serious complications that respondents could recall: 14 incidences of respiratory depression, 6 of delayed sedation and 2 of hypotension were reported. Only two of the reported adverse events occurred in hospitals using doses of up to 1 mg of diamorphine. There was no stratification asked for to define respiratory depression, e.g. need for reversal with naloxone, need for ventilation, etc. There was no reported mortality associated with intrathecal opioid usage.
Association of Anaesthetists of Great Britain and Ireland (AAGBI) guidelines for intrathecal opioid usage
In all, 109 (55.1%) respondents thought that the Association guidance would be either very useful or useful, 33 (16.7%) were undecided and 16 (8.1%) did not feel that guidance would be of use; the remaining 40 (20.2%) gave no response. The percentages were almost identical in those departments with pre-existing guidelines and those without.
Of the six paediatric hospitals returning the survey, only one used intrathecal opioids and it did not have guidelines for their use. Diamorphine was used in spinal surgery and the patients went to ICU postoperatively. The nurses received verbal post-anaesthetic instructions and the patients were then put on normal opioid prescriptions with care. The department had not had any serious complications or adverse events.
This questionnaire attempted to obtain a snapshot of intrathecal opioid use by anaesthetic departments in the UK. The results of this survey need to be interpreted with caution as the information given was retrospective, and may suffer from recall bias and therefore underestimate the actual occurrence rate of adverse events. The return rate of 73.7%, although good, may have subjected the results to bias by departments keen on intrathecal opioid use (non-responders may have more unusual practices). However, much of the data are exact, such as opioids used and dosages, destination of the patients and postoperative guidelines.
The questionnaire confirmed the widespread use of a variety of intrathecal opioids, with diamorphine in 136 (78.2%) departments and fentanyl in 129 (74.1%) superseding morphine (37, 21.3%), which was the most commonly used opioid in previous surveys [2,3]. Morphine is still the only licensed opioid for intrathecal administration. No other opioid has undergone neurotoxicity studies in humans to establish safety. Some older opioid preparations contained preservatives which, if injected intrathecally or epidurally, may increase the risk of nerve damage [4,5].
The doses of opioids used were generally similar although there were notable exceptions. Most departments used morphine 0.1-0.5 mg (62.2% of departments who routinely used morphine as an intrathecal opioid), diamorphine 0.2-0.5 mg (67.6%) and fentanyl 12.5-25 μg (70.5%). Seven departments reported using more than 1 mg of diamorphine with two using doses of 3.5 and 4 mg, respectively, without reporting any adverse events. Most adverse events occurred in departments using average doses of opioids. Two departments which had no guidelines and which reported an adverse event used diamorphine in doses of up to 1 mg routinely. Alfentanil, meperidine and buprenorphine were also used in four different departments. No guidelines were in place regarding these drugs, and only the buprenorphine dose was given by the respondents. These departments did not report any adverse events.
The level of post-anaesthetic care for patients who have received spinal opioids has been debated since 1982 . Morgan concluded that the use of intrathecal and epidural opioids was impracticable for the majority of patients, due to the close postoperative surveillance necessary to monitor respiratory depression and other complications. In 1989, Morgan  reviewed this topic and concluded that any patient undergoing either of these forms of analgesia needed monitoring on either an ICU or HDU for 24 h. Morphine was the most commonly used drug at that time and often was given in doses exceeding 1 mg. Morphine is highly hydrophilic and enters the spinal cord slowly, producing persistently high cerebrospinal fluid (CSF) concentrations [8,9], which allows significant cephalad migration. Observations of CSF concentrations have shown that it takes 1-2 h for morphine to reach the cisterna magna from the lumbar region and 4-8 h to pass through the foramina of Luschka and Magendie to the fourth ventricle . Subsequent interaction with high concentrations of opioid receptors located in the ventral medulla can result in delayed respiratory depression [11-13]. Factors that accelerate the circulation of CSF such as coughing, sneezing or straining may result in a more rapid appearance of symptoms and a more profound effect. Other factors have been noted to predispose to respiratory depression, e.g. advanced age, concomitant use of parenteral opioids or other CNS depressants, and a lack of tolerance to opioids. Unfortunately, the term ‘respiratory depression' has no clear definition [14,15], and although using bradypnoea is useful it can be inadequate as hypercapnia may develop despite a normal respiratory rate . Pulse oximetry monitoring may be useful  but needs cautious interpretation if supplemental oxygen is being administered .
A depressed level of consciousness, possibly caused by hypercapnia, may be the most reliable sign of respiratory depression [16,19]. Respiratory depression was recalled by six departments in the survey, all of which had guidelines in place for the use of several opioids. The historical incidence of respiratory depression is variable [20-22] with quoted figures as high as 3%  and 4-7%  in earlier studies, down to 0.36% and 0.03% in more recent Swedish surveys [2,3]. This may be due to the use of lower doses and the avoidance of concomitant sedatives and opioids by other routes. This more recent lower incidence of respiratory depression compares favourably with the 1.2% incidence quoted for opioid administration via PCA  and with the 0.9% incidence with the use of intramuscular (i.m.) opioids . Data from over 20 000 patients show that the respiratory depression manifests within 12 h following intrathecal morphine injection [2,3]. For lipophilic drugs, e.g. fentanyl or diamorphine, the danger period could be as low as 4-6 h after administration, although more data are required before safe observation routines can be established.
Urinary retention is a common side-effect particularly in young males having intrathecal morphine [2,27], independent of the dose  being administered. This complication was mentioned only by one department which used only long-acting opioids, but this may be due to urinary retention not being a pre-fixed answer choice on the questionnaire proforma.
The Swedish Society of Anesthesiology and Intensive Care published guidelines for the post-anaesthetic monitoring of patients receiving central neuraxial opioids in 1992. They state that patients can be nursed safely on regular wards if: (1) personnel are trained and if pre-printed guidelines for potential emergencies are provided; (2) patient selection and opioid dosing are appropriate; and (3) respiratory rate and level of sedation are checked every hour for 12 h. Similar guidelines have been recommended by the European Society of Regional Anaesthesia for epidural opioid usage . A survey of 17 European Nations hospital practices  showed that 60% of hospitals nursed their patients on regular surgical wards. Also, 26% of European and 40% of UK hospitals nursed ASA III-IV patients on regular wards following intrathecal opioid administration. In our survey, the majority of patients went either to a normal ward (in 96 (54.5%) departments) or to a variable location including ward/HDU in 46 (26.1%). There was a tendency for departments with guidelines to send more of their patients to a critical care area postoperatively (14.2% vs. 7.4%).
Further doses of opioids during the first postoperative day are an area of concern. Our departmental guidelines dictate that all patients are connected to a PCA morphine system for 24 h. This means that regular observations (including respiratory rate and sedation scores) are carried out and no other sedatives or opioids may be given. This routine is already familiar to the nursing staff and was used by 22 (12.6%) of the departments in the survey. The most common method of prescribing was delaying any opioid prescription for 24 h after administration of intrathecal opioid.
Post-anaesthetic monitoring protocols for nursing staff were available in 54% of European hospitals  compared to 56.9% of hospitals in this survey. An additional 23.5% of departments gave some written guidance postoperatively and 27.6% gave some form of verbal information. Only 12.6% of departments gave no post-anaesthetic instructions to the medical/nursing staff.
Of the departments, 55% thought that AAGBI guidelines would be useful but 8% were opposed to further guidelines that may restrict their practices.
In conclusion, this survey has illustrated that intrathecal opioid administration is widespread in UK anaesthetic practice. Diamorphine (0.1-0.5 mg) and fentanyl (12.5-25 μg) are now the most commonly used opioids. Patients are sent to regular surgical wards in 54.5% of departments and the nursing staff receives either written or verbal instruction on observations in 84.6%. Normal opioids are given with care in 19.5% of departments or there is a delay in prescribing further opioids in 28.7%. The incidence of major complications associated with intrathecal opioids, that can be recalled, is minimal. This is probably a reflection of a dose reduction and the increasing popularity of lipophilic opioids (diamorphine and fentanyl). Postoperative care protocols to avoid large doses of opioid being administered by another route and any sedatives being given may also have contributed to a safety profile comparable to (if not better than) PCA and i.m. administration.
As 55% of returnees stated that AAGBI guidelines would be useful for their departments, it may now be time to draw up national guidelines for intrathecal opioid administration to further standardize their administration and improve patient safety.
Many thanks to Reza Veziri and Dr Jan Jones at the QMC Audit Department for formatting the survey and collating the information received.
1. Racoviceanu-Pitesti N. Anesthesia generala prin cocaina. Rev Chir
1900; IV: 11-22.
2. Rawal N, Arner S, Gustafsson L, Allvin R. Present state of extradural and intrathecal opioid analgesia in Sweden. Br J Anaesth
1987; 59: 791-799.
3. Rawal N, Allvin R. Acute pain services in Europe: a 17-nation survey of 105 hospitals. Eur J Anaesthesiol
1998; 15(3): 354-363.
4. Du Pen S, Ramsey D, Chin S. Chronic epidural morphine and preservative-induced injury. Anesthesiology
1987; 67: 987-988.
5. Mathews E. Epidural morphine (Letter). Lancet
1979; 1: 673.
6. Morgan M. Intrathecal and epidural opiates: an assessment. Anaesthesia
1982; 37: 1365-1366.
7. Morgan M. The rational use of intrathecal and epidural opioids. Br J Anaesth
1989; 63: 165-188.
8. Nordberg G, Hedner T, Mellstrand T, Dahlström B. Pharmacokinetic aspects of intrathecal morphine analgesia. Anesthesiology
1984; 60: 448-454.
9. Sjöström S, Tamsen A, Persson M, Hartvig P. Pharmacokinetics of intrathecal morphine and meperidine in humans. Anesthesiology
1987; 67: 889-895.
10. DiChiro G. Observation on the circulation of the cerebrospinal fluid. Acta Radiol Diagn
1966; 5: 988-1002.
11. Shook J, Watkins W, Camporesi E. Differential roles of opioid receptors in respiration, respiratory disease, and opiate-induced respiratory depression. Am Rev Respir Dis
1990; 142: 895-909.
12. Mueller R, Lundberg D, Breese G, Hedner J, Hedner T, Jonason J. The neuropharmacology of respiratory control. Pharmacol Rev
1982; 34: 255-285.
13. Gregory M, Brock-Utne J, Bux S, Downing J. Morphine concentration in brain and spinal cord after subarachnoid morphine injection in baboons. Anesth Analg
1985; 64: 929-932.
14. Jordan C. Assessment of the effects of drugs on respiration. Br J Anaesth
1982; 54: 763-782.
15. Ko S, Goldstein DH, VanDenKerkhof EG. Definitions of ‘respiratory depression' with intrathecal morphine postoperative analgesia: a review of the literature. Can J Anaesth
2003; 50(7): 679-688.
16. Ready L, Oden R, Chadwick H et al
. Development of an anesthesiology-based postoperative pain management service. Anesthesiology
1988; 68: 100-106.
17. Bailey P, Rhondeau S, Schafer P et al
. Dose-response pharmacology of intrathecal morphine in human volunteers. Anesthesiology
1993; 79: 49-59.
18. Chaney M. Pulse oximetry during conscious sedation (Letter). JAMA
1994; 271: 429.
19. Etches RC, Sandler AN, Daley MD. Respiratory depression and spinal opioids [see comment]. Can J Anaesth
1989; 36(2): 165-185.
20. Gustafsson L, Schildt B, Jacobsen K. Adverse effects of extradural and intrathecal opiates: report of a nationwide survey in Sweden. Br J Anaesth
1982; 54: 479-486.
21. Palmer C. Early respiratory depression following intrathecal fentanyl-morphine combination. Anesthesiology
1991; 74: 1153-1155.
22. Hays R, Palmer C. Respiratory depression after intrathecal sufentanil during labor. Anesthesiology
1994; 81: 511-512.
23. Gwirtz K, Young J, Byers R et al
. The safety and efficacy of intrathecal opioid analgesia for acute postoperative pain: seven years' experience with 5969 surgical patients at Indiana University Hospital. Anesth Analg
1999; 88: 599.
24. Cousins M, Mather E. Intrathecal and epidural administration of opioids. Anesthesiology
1984; 61: 276-310.
25. Coleman J, Dolin S. Respiratory and haemodynamic effects of acute postoperative pain management: evidence from published data. Br J Anaesth
2004; 93: 212-223.
26. Miller R, Greenblatt D. Drug effects in hospital patients. In: Miller RR, Greenblatt DJ, eds. Experiences of the Boston Collaborative Drug Surveillance Program, 1966-1975
. New York: John Wiley and Sons, 1976.
27. Bromage P, Camporesi E, Durant P, Nielson C. Nonrespiratory side effects of epidural morphine. Anesth Analg
1982; 61: 490-495.
28. Aguilar J, Benhamou D, Bonnet F et al
. ESRA guidelines
for the use of epidural opioids. Int Monitor Reg Anaesth
1997; 9: 3-8.