Rationalising the use of radiology services-Increasing the benefit, reducing the risk

Summary

Throughout the world, a large number of x-ray pathology suspected or do not alter the management of the examinations are requested which are of little or no clinical patient value. These examinations are either inappropriate for the pathology suspected or do not alter the management of the patient.

Introduction

The World Health Organisation (WHO) in their publication “A Rational Approach to Radiodiagnostic Investigations”¹ says: - “There needs to be more rational use of scarce and over-utilised radiological facilities. Where resources are limited, more careful selection of patients will make the benefits of diagnostic radiology available to those in need.”

Radiographs are expensive to produce - radiology departments are known to absorb a large slice of hospital budgets. Healthcare resources are finite in all countries, but in Malawi they are extremely limited. In 1998, the last year for which figures are available, spending on health care amounted to only US $5 per person per year.² Even simple x-ray investigations can cost upwards of $10 so it is clear that if there really is to be ‘health for all’ in the country then some sort of prioritisation of needs must take place.

Producing radiographs also involves the use of ionising radiation, which carries with it a small but significant risk to the patient. Ionising radiation has known links with a range of malignancies (leukemia, thyroid, breast, bone and lung cancers) and genetic conditions (Huntingtons Chorea, Cystic Fibrosis, Tay Sachs disorder, Haemophilia and Albinism).³ The risks of radiation are stochastic (random in nature) and non-threshold (meaning there is no ‘safe’ limit). In deciding whether to refer a patient for x-ray, clinicians must balance the benefits of performing a given test against the risks involved in that test. Even in a hypothetical situation of infinite resources, therefore, it would still be necessary to limit x-ray requests to circumstances where there would be a clear clinical advantage to be gained.

Reducing the number of unnecessary or inappropriate x-ray examinations carried out will therefore bring a number of benefits. The risk of radiation-induced pathologies will be lessened and money saved can be spent on improving patient care in other areas. In addition, waiting times for clinically justified investigations will be reduced and staff time (doctors, CO's nurses and radiographers) is saved for more productive activities.

Before a patient is referred to the radiology department, a number of points should be considered:-

• Has an appropriate history been taken and detailed clinical examination done?

• Will the examination requested actually demonstrate the pathology suspected?

• Is this the most appropriate examination available?

• Is there someone trained who can interpret the films?

• Will the results alter the patient management?

Below are a number of examples in order to illustrate these points. They are not intended to be ‘hard and fast’ rules and cannot be applied blindly to every case, nor by any means are they the only areas in which radiology referrals may be rationalised. They may however raise some points for discussion.

Example 1a-Abdominal X-Rays in Suspected Bowel Obstruction

The symptoms and signs of bowel obstruction are usually very specifie (see table below - information taken from Common Medical Problems in Malawi.” The symptoms and signs of paralytic illeus are given for comparison.

Intestinal Obstruction	Paralytic Illeus
Colicky Abdomen Pain	No Colicky Pain
Abdomen Distension	Bowel Sounds Absent
Vomiting
Constipation
Visible Peristalsis
Tinkling Bowel Sounds

A study undertaken in Germany⁵ followed 1254 patients with acute abdominal pain lasting less than 7 days and with no recent history of abdominal injury or surgery.⁶ common findings were identified, namely distended abdomen, increased bowel sounds, history of constipation, previous abdominal surgery, age over 50 and vomiting. The study found that if only patients presenting with any two (or more) of these symptoms had had radiographs taken then 42.6% could have been avoided without loss in diagnostic accuracy.

Conversely, abdominal x-rays in patients with a strong clinical diagnosis of intestinal obstruction may be unnecessary and simply delay appropriate treatment of the patient.

Summary

“X-rays are often ordered in patients suspected of intestinal obstruction but do not really alter the management, which should be based on clinical assessment.”

Common Medical Problems in Malawi⁴

Example 1b - Erect Abdomen X-Rays in the Acute Abdomen

Whilst a significant number of abdominal x-rays for intestinal obstruction might be avoided, radiographs will still be required where the diagnosis remains equivocal. In such cases a 3 film series is often requested, consisting of a supine and an erect abdominal radiograph and an erect chest x-ray.

Two reasons are normally given for requesting erect abdominal x-rays - to demonstrate sub-diaphragmatic free air in suspected bowel perforation and to demonstrate fluid levels in cases of suspected bowel obstruction. However, a number of studies have suggested that there is very little indication for erect views of the acute abdomen.^6,7,8

Due to a number of technical considerations, sub-diaphragmatic air from bowel perforation is significantly better demonstrated on an erect chest x-ray than an erect abdomen. It is important to ensure that the patient has been erect for at least 10 minutes before x-ray is taken in order to allow time for gas to rise. If this is not possible due to the patients' condition, a left lateral decubitus abdomen will clearly demonstrate any free air surrounding the liver. An erect chest x-ray will also demonstrate some intrathoracic conditions that may present as acute abdominal pain and mimic intestinal obstruction or perforation.

Multiple fluid levels can be observed in the normal, asymptomatic erect abdomen x-ray.⁹ In addition, ileus, ischaemia, gastroenteritis and diabetes may cause fluid levels. This then raises the diagnostic dilemma of whether fluid levels seen are the result of bowel obstruction or not. An erect abdomen therefore cannot be used to either confirm or exclude a diagnosis of obstruction. Fluid levels also give no indication as to the level of any obstruction present.⁶

It is suggested therefore that where radiographic examination of the acute abdomen is required, this should be limited to a supine view of the abdomen and an erect view of the chest.

Summary

“The addition of the erect film does not significantly increase the accuracy of diagnosis of obstruction or help in the correct identification of its level.”

The Value of an Erect Abdominal Radiograph in the Diagnosis of Intestinal Obstruction Clinical Radiology⁶

Example 2 - Chest X-Rays for Simple Rib Fractures

Patients frequently present in the Out-patients department (OPD) with localised chest pain following traum. The patient will first be examined clinically for breathlessness, haemoptysis, abnormal breath sounds etc. If any of these are discovered the patient should be referred for radiographs in order to confirm and determine the extent of the injury. In the absence of clinical signs however, it is very unlikely that there is significant injury to the underlying lung.

It is common for the patient then to be referred for x-ray to demonstrate a simple rib fracture.

In the event that a fractured rib is demonstrated, but in the absence of other underlying injuries, patients are likely to be advised to rest and mild analgesics may be prescribed. The same treatment is likely to be recommended for a patient with soft tissue bruising but no demonstrated fracture.

The radiographic demonstration of a rib fracture in the absence of clinical signs of pleural injury therefore adds little to the management of the patient.

In addition, it is important to note that simple chest x-rays are very unreliable at demonstrating simple rib fractures. A chest radiograph that appears ‘clear’ therefore cannot exclude a fracture.

Summary

“Chest radiography for chest trauma is not indicated unless secondary complications are suspected. The radiological demonstration of simple rib fractures is unreliable and unlikely to alter treatment.”

World Health Organisation

A Rational Approach to Radiodiagnostic Investigations¹⁰

Example 3 - Skull X-Rays Following Head Injury

A similar argument can be put forward with regard to most skuU x-rays following trauma.

A skull radiograph can only demonstrate the bony skuU; it does not show the soft tissues of the brain or meninges. The presence of a skull fracture may increase the possibility of intracranial injury, but many patients have intracranial injury without skull fracture.¹¹ One study of 1845 patients with head injury reported 33 patients with significant intracranial heamorrhage. Of these, only 7 had a skull fracture.¹²

A displaced calcified pineal gland may raise suspicion of intracranial bleeding but this is a rare finding, especially outside the older age groups. Signs of intracranial heamorrhage are therefore better detected by clinical examination.

Since there is no specific treatment for non-depressed skull fractures, their demonstration on radiographs adds nothing to the management of the patient and may be considered unnecessary.

Intracranial haemorrahage may only be demonstrated by more sophisticated imaging methods such as Computed Tomography (CT scanning). These are expensive and result in high radiation doses. However, where they are available they may prove useful in the patient with clinical signs that are suggestive of intracranial injury who is likely to benefit from surgery.

A depressed skull fracture may be suspected clinically. Significant intracranial injury is then likely and a skull radiograph may be required to determine the position of the depressed fragments with a view to possible surgery.

Summary

“Skull radiography is indicated only where there is clinical suspicion of a depressed fracture. In the absence of such a fracture, 98% of skull radiographs do not influence patient management or outcome. The discovery of such a linear skull fracture is of very limited clinical significance.”

World Health Organisation

Effective Choices for Diagnostic Imaging in Clinical Practice¹

Example 4 - Management of Non-Traumatic Low Back Pain

Low back pain is an extremely common presenting symptom - in developed countries between 70% and 90% of people experience low back pain at some point in their lives^13,14 and its prevalence also appears to be high in Malawi. Pain is non-specific in the majority (about 85%) of sufferers. Non-specific low back pain is low back pain not attributed to recognisable pathology (such as infection, tumor, osteoporosis, rheumatoid arthritis, fracture or inflammation).

The role of imaging is to investigate those patients who show warning signs of conditions where aggressive treatment is both required and obtainable such as tuberculosis, and possibly neoplasms.

Warning Signs in Non-Traumatic Back Pain

• < 20 years or >55

• Alternating or bilateral sciatica

• Weak legs

• Signs of systemic illness (weight loss, pyrexia of unknown origin, raised erythrocyte sedimentation rate, HIV)

• Systemic steroid use

• Progressive, non-mechanical back pain (Unremitting, Unrelieved by Rest)

• Spine movements in all directions painful

• Localised bony tenderness

• CNS deficit of more than one root level

• Pain or tenderness of the thoracic spine

• Bilateral signs of nerve root tension

• Past history or suspicion of neoplasia

Adapted front Oxford Handbook of Clinical Medicine¹⁵

Patients not showing warning signs are likely to be suffering from non-specific low back pain. This can occur as a result of a number of benign causes. Inflammation of facet joints, capsules, spinous ligaments and intervertebral disks can all cause pain. Often there is spasm in the paraspinal muscles exacerbating the pain. It is difficult to distinguish these etiologies clinically. Fortunately, conservative therapy is usually sufficient for all of these, making such a distinction not cntical.¹⁴ If the pain is acute they are likely to benefit from advice to stay active and analgesics. If chronic, back exercises, non-steroidal anti-inflammatory drugs (NSAIDS) and back care education may help.¹⁶ These patients do not normally require further investigation.

It should be borne in mind that degenerative changes are almost always present on spinal radiographs of older patients whether symptomatic or not - by age 50. more than 80% of adults demonstrate radiographic evidence of spondylosis.¹⁷ Demonstration of degenerative changes in these patients is therefore of little clinical significance.

Summary

“Studies have shown that imaging of asymptomatic individuals will reveal a relatively high percentage of anatomie abnormalities that undersome circumstances cause symptoms. Therefore, one should not do any imaging [of the spine] unless there is strong clinical suspicion of disease with neurologie findings present.”

Imaging Handbook for House Officers¹⁸”

Conclusion

Most developing countries have extremely limited health resources. By rationalising the use of radiology services the clinician can increase the benefits in terms of value for money whilst at the same time reducing the risk to the patient from ionising radiation.