Intended for healthcare professionals

Clinical Review Lesson of the week

Postoperative pressure sores after epidural anaesthesia

BMJ 2000; 321 doi: (Published 14 October 2000) Cite this as: BMJ 2000;321:941
  1. J L Shah, consultant (JantiShah{at}
  1. Department of Anaesthetics, City Hospital, Birmingham B18 7QH
  1. Correspondence to: J L Shah, Department of Anaesthetics, Queen Elizabeth Hospital, Edgbaston, Birmingham B15 2TH
  • Accepted 21 January 2000

Pressure sores may occur when epidural analgesia produces profound sensory and motor block

Epidural anaesthesia and analgesia provide a better outcome after major surgery.1 Side effects of epidural such as dural puncture, motor block, and hypotension are well known. Epidural analgesia may also cause pressure sores in elderly and debilitated patients.2 The general belief is that pressure sores do not occur in patients who are fit and mobile. This report describes three cases of heel ulcers in fit young patients related to pressure after epidural analgesia. None of the patients had any predisposition to pressure related ulceration. In two of the patients the heel ulcers became apparent several days after surgery.

Case reports

Case 1—A 52 year old woman weighing 42 kg had carcinoma of the vulva. She was otherwise fit and active. She smoked 20 cigarettes a day. She had a radical vulvectomy and bilateral inguinal lymphadenectomy under combined general and epidural anaesthesia. She was given 20 ml of 0.25% bupivacaine for epidural anaesthesia. To prevent pressure sores, the operating table was covered with silicon jelly pads. Surgery lasted about three hours. The patient was supine for the first 90 minutes and in the lithotomy position thereafter. Her preoperative blood pressure was 120/70 mm Hg. During surgery her systolic blood pressure varied between 85 and 90 mm Hg and her general condition was stable. She received a continuous epidural infusion of plain 0.15% bupivacaine for postoperative analgesia. She remained free of pain and was comfortable. Her systolic blood pressure varied between 75 and 85 mm Hg. She was unable to move her legs on the first postoperative day, but could move them on the second day. On the third day, the epidural was discontinued, and she could get out of bed and walk. On the fourth day she noticed blisters and small areas of discoloration on her heels. Over the next three days the blisters developed into ulcers. Five weeks later at outpatient follow up her heels were worse, with severe pressure necrosis of both heels (fig 1). She had lymphoedema of both legs. The ulcers took between eight and nine months to heal.

Fig 1.
Fig 1.

Heel ulcers in both feet of case 1. Reproduced with patient's permission

Case 2—A 35 year old fit and active woman weighing 63 kg had a Wertheim's hysterectomy for carcinoma of the cervix. She was given 20 ml of 0.125% bupivacaine for epidural anaesthesia. She was supine on silicon jelly pads on the operating table for two and a half hours. Her preoperative blood pressure was 110/60 mm Hg. During surgery her systolic blood pressure varied between 105 and 110 mm Hg. After surgery she was free of pain and able to move both legs. She received a continuous epidural infusion of plain 0.2% ropivacaine for postoperative analgesia. She remained free of pain and was able to move both legs. Her postoperative systolic blood pressure remained above 95 mm Hg. On the second day she complained of sore heels. The heels were raised off the mattress and she was advised to change the position of her legs regularly. The epidural was discontinued on the morning of the third postoperative day, and she was able to get out of bed. She was discharged home without further apparent problems. Two weeks later at outpatient follow up she had ulcers on both heels (fig 2). The ulcers took six weeks to heal.

Case 3—A 33 year old fit and healthy woman weighing 80 kg had a Wertheim's hysterectomy for carcinoma of the cervix. Her blood pressure was 130/76 mm Hg. During the three and a half hours of surgery, her systolic blood pressure remained above 95 mm Hg. She received a continuous epidural infusion of 0.15% bupivacaine with 20 µg/ml diamorphine for postoperative analgesia. Her systolic blood pressure remained above 100 mm Hg. On the second postoperative day her right leg was numb and she could not move it. On the third postoperative day, after discontinuation of her epidural, she was able to move normally. She made a good recovery and was discharged home. Three weeks later at outpatient follow up she had a discoloured right heel with a pressure sore. It took a further four weeks for the ulcer to heal.


The risk of pressure sores is greatest in patients who are elderly, debilitated, incontinent, paralysed, or unconscious and in those with metastatic cancer. 3 4 The patients in this report were young, fit, and well hydrated, and they all had early stages of cancer. Carcinoma of the cervix may be found at routine cervical smear screening in otherwise fit women. Under normal circumstances these patients would not be expected to develop pressure sores.

The surgery in our patients was not prolonged, they did not have excessive hypotension, and they were kept on silicon jelly pads to relieve any pressure during surgery. It is therefore unlikely that the damage to their heels occurred during surgery.

Epidural analgesia produces sensory and motor block, and it restricts a patient's movement. Restricted movement of the legs may result in prolonged pressure on the heels. A constant pressure of 70 mm Hg for more than two hours produces tissue ischaemia and irreversible tissue damage.5 Case 3 had a unilateral motor and sensory loss, and she developed a heel ulcer on the blocked side only.

Fig 2.
Fig 2.

Heel ulcers in both feet of case 2. Reproduced with patient's permission

Laser Doppler fluxometry is a non-invasive technique for measuring blood flow in skin. A laser light applied to the skin penetrates to a depth of 1 mm. The light reflects back to a detector. The comparison of reflected light between the static skin and moving blood gives a qualitative estimate of blood flow. This technique shows a great reduction in blood flow in the skin when a patient lies on a conventional NHS bed.6 Patients at increased risk of pressure sores should therefore be nursed on pressure relieving mattresses such as the large celled ripple bed 7 8; none of our patients were nursed on such mattresses.

Occlusion of lymphatic vessels produces an increase in tissue pressure and oedema, and it also leads to accumulation of anaerobic metabolic waste products.9 Bilateral inguinal lymphadenectomy in Case 1 produced severe oedema of her legs. This may have exacerbated the ulceration of her heel and delayed the healing response.

Previous reports have shown that heel damage may occur within 24 hours after spinal and epidural analgesia. 10 11 It is likely that our patients' heels were damaged on the first postoperative day. In the early stages the damage may go unnoticed by patients and healthcare staff; the ulcers may appear several days after the initial pressure injury when patients are likely to have been discharged from hospital. Ulcers may heal by the time patients attend for outpatient follow up and therefore the injury goes unreported. As such it is difficult to assess the prevalence of this rare and unusual complication of epidural anaesthesia and analgesia.

Epidural analgesia represents a major advance in the care of patients undergoing major surgery. All staff who care for patients receiving epidural analgesia should be alert to the potentially serious problem of pressure related ulcers. Assessment for such problems should be a routine part of monitoring patients receiving epidural analgesia. The patients should be nursed on pressure relieving mattresses, and they should wear heel pads. Patients with motor block should be turned regularly.

To prevent pressure sores in patients receiving epidural analgesia heel pads should be worn routinely, recovery should take place on large celled ripple type mattresses, hypotension should be avoided, patients with motor block should be treated as paraplegic patients and turned regularly, and motor and sensory block should be avoided in legs (lower thoracic instead of lumbar epidural block should be considered).


  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
View Abstract

Log in

Log in through your institution


* For online subscription