Case 16 Back pain in a middle-aged woman

Angela had been registered at the practice for three years when she consulted Dr Ahmed about her back pain. She had initially become aware of lumbar ache four months earlier, after helping her daughter move into a new flat. The pain was just to the right of the lower thoracic spine with no radiation. Initially the pain had been relieved by ibuprofen or paracetamol but more recently the pain was more persistent and was disturbing her sleep somewhat. She was otherwise well.

What would you do now? Are any particular features on examination likely to be helpful?

Dr Ahmed found that Angela had a full range of movement of the spine and straight leg raising was symmetric and unimpeded. Spinal percussion was not painful.

Dr Ahemd prescribed some co-codamol, advised Angela to keep mobile and suggested she return if the pain was no better in one month.

A month later Angela returned and the pain was no better. She described the pain as being slightly lower down the back by this time, and she had pain radiating into her right anterior thigh which kept her awake at night.

What would be your differential diagnosis and how would you discriminate between them?

Dr Ahmed found that there was no weakness of hip flexion or knee flexion and extension but that right ankle dorsiflexion was weak and the right knee reflex was absent. He wondered about a lumbar disc prolapse but decided to obtain blood tests (FBC, ESR, CRP, U+E and LFTs) and obtain X-rays of the lumbar spine.

The ESR was 28 but otherwise the blood tests were largely normal. However, the X-ray was reported as showing a bone cyst in the third lumbar vertebra and the radiologist’s report suggested a myeloma screen. Dr Ahmed asked the receptionists to ask Angela to make an appointment for a blood and urine test. He was then on holiday for two weeks. The myeloma screen was seen by one of his colleagues and was negative.

Three months later Angela returned to see one of Dr Ahmed’s colleagues. She was in a great deal of pain and had lost some weight. Her right leg was weak and she had difficulty passing urine. She was admitted to hospital and found to have a large vertebral metastasis affecting L3 and L4 and cord compression. It was noted in hospital that she had a history of breast cancer treated successfully 18 years earlier. This information was in the general practice correspondence but was not on the computerized summary. She underwent emergency radiotherapy but was left with urinary incontinence and weakness of her right leg.

Angela brought a claim against Dr Ahmed alleging that he should have considered her past history of breast cancer, her night pain and her age and referred her for urgent specialist care. It was further alleged that the lumbar spine X-ray results had not been acted upon.

Do you think her claim will succeed?

Expert comment

Back pain is extremely common in the general population and is an extremely common complaint in general practice. The annual incidence of significant back pain is estimated at 5% and the point prevalence in the population at 25%. Most back pain is benign and self-limiting. 90% of people will have substantially improved by 6 weeks from the onset.

In primary care approximately 0.7% of patients presenting with back pain have metastatic cancer, 4% have osteoporotic stress fractures, 0.3% have ankylosing spondylitis and 0.01% have spinal infections (Deyo et al., 1992; Jarvik & Deyp, 2002).

The 1999 guidance document produced by the Royal College of General Practitioners listed the following ‘red flags’ (with referral guidance):

Red flags for possible serious spinal pathology:

• Presentation under age 20 or onset over 55
  
• Non-mechanical pain
  
• Thoracic pain
  
• Past history – carcinoma, steroids, HIV
  
• Unwell, weight loss
  
• Widespread neurological symptoms or signs
  
• Structural deformity

The difficulty is that not all these ‘red flags’ are equal. Studies show that some, such as age over 55, nonmechanical back pain and thoracic back pain, make up a significant proportion of the population consulting in primary care and are not particularly helpful for identifying rare cases of cancer or infection. Most studies indicate that back pain is not relieved at night by lying down in nearly half of all patients with simple mechanical back pain (Deyo et al., 1992; Van den Hoogen et al., 1995). Nearly a sixth of people consulting have thoracic pain and one-third to one-quarter are aged over 55 (Deyo et al., 1992; Van den Hoogen et al., 1995).

However, a history of cancer, particularly those cancers that tend to metastasize to bone (such as breast, colon, prostate, lung, renal) has a very high likelihood that the back pain will be due to metastases. In one study of 1975 patients presenting in primary care 13 had bone metastases. Out of the 1975 only 45 had a history of cancer and 4 (9%) had vertebral metastases causing their back pain (Deyo et al., 1992). Other studies also suggest that about 10% for of those with back and a history of cancer will have bone metastases.

Spinal tenderness to percussion is equally common in patients with and without cancer and is unhelpful.

An ESR of over 20 mm/hr occurs in about 80% of patients with vertebral metastases. However, the usefulness of the test is limited by the fact that about a third of patients without cancer have an ESR over 20 (the specificity is low at about 65%). 70% of patients with vertebral metastases have X-ray changes (lytic or sclerotic lesions or crush fractures). Virtually no patients without cancer or a vertebral crush fracture will have these clinical features, making an X-ray reasonably discriminatory though isotope bone scans are obviously much more sensitive.

This case does illustrate several problems that are not uncommon in medico-legal cases.

It is clinically unsafe if significant past medical events are not recorded in a prominent position in computerized notes. In this case the past history of treated breast cancer was clear in the correspondence, but when the notes were being summarized for the computer when Angela registered at the practice the history was not recorded on the summary. This can lead to very serious clinical oversights, as in this case. A junior doctor seeing a patient for the first time in hospital will usually ask the patient about their past medical history. A general practitioner rarely does this in a 10-minute consultation if there appears to be a complete summary on the computer.

A second relatively common problem occurs when another general practitioner sees a normal result, is not familiar with the patient but assumes that, because the result is normal (the myeloma screen in this case) no further action is necessary. It is understandable that the colleague checking Dr Ahmed’s results does not scrutinize the history with all normal blood results. However, Dr Ahmed needed to alert the colleague covering in his absence that Angela had a lytic bone lesion, and that further action was required.

A third problem is that general practitioners are not used to looking at or interpreting X-rays. As such they rely heavily on the radiologist’s report. In this case an expert radiologist was critical of the standard of the radiology report. It did not make it clear to the general practitioner that the differential diagnosis of the lytic bone lesion included things other than myeloma. However, it was reasonable to expect a general practitioner to know this.

Dr Ahmed had considered that the likely pathology was a disc prolapse affecting the right L4 nerve (and causing the loss of the knee reflex). However, disc prolapses other than L5/S1 or L4/L5 are rather unusual and, particularly in the older patient, should but the general practitioner on alert that the radiculopathy may be due to malignant disease.

The case was clearly indefensible on breach of duty. The practice as a whole was liable for the poor quality of the computerized notes and Dr Ahmed was liable for failing to act in the light of the past history of cancer and failing to act properly on the radiology report. The case illustrates how medical errors normally only occur when several things go wrong sequentially.

Legal comment

Expert opinion is clear that there has been a breach of duty by Dr Ahmed. It seems that from the start he should have had the possibility in mind of a metastasis. As it was, the diagnosis was made about four months late. If Angela had had radiotherapy four months sooner, what would the outcome have been on the balance of probabilities? Would she have had urinary incontinence and/or weakness of the right leg?

Angela’s lawyers and Dr Ahmed’s lawyers will each obtain an expert opinion on these causation questions. They may well arrange for these two experts to meet to discuss the case, and see if they can come to a common view. If they conclude that earlier treatment would have saved Angela from the leg weakness and incontinence, the lawyers will start to negotiate a settlement.

It may be possible to obtain a contribution towards the settlement from the hospital in relation to the poor quality of the X-ray report. If the practice is comprised of partners who belong to a different MDO from that of Dr Ahmed, then his MDO may also ask for a contribution from them in relation to the practice’s failure to keep proper computerized records. However, these contributions are not likely to be significant.

Stay updated, free articles. Join our Telegram channel

Join