Acute abdominal pain in general practice

GP Jonathan Berry considers when to refer in acute abdominal pain 

AS a hospital junior over 30 years ago, working in a busy general surgical and then gynaecology unit, I had quite good diagnostic skills with regard to acute abdominal pain. We rarely took out a “normal” appendix, and never operated for a probable ectopic pregnancy without finding some significant acute gynaecological problem requiring surgery, even if not an ectopic. A careful history, examination and simple tests, such as urine dip and a full blood count, almost invariably gave the correct answer. And it was easy to observe a patient for a few hours if there was uncertainty and we never sent home a patient who returned seriously ill within a few days.

Familiarity with the common acute abdominal presentations was probably the key. We didn’t rely on diagnostic tests and there was no on-call ultrasound scanning in any event. A few years after becoming a GP, I was amused to receive a discharge letter critical of my having admitted a patient as a possible appendicitis. The final paragraph stated that histologically the appendix was normal! I don’t think I was wrong to admit. Was an appendicectomy warranted? Well, the operating surgeon thought so.

Some stats

In today’s NHS we are rightly asked to reduce waste and especially to cut down on unnecessary unscheduled care admissions. A recent study by Brekke and Eilerston* found that of 26 per cent of patients seen in general practice with acute abdominal pain and admitted, nearly half had appendicitis and many of the remainder had gyaenacological, biliary tract or peptic ulcer disease. Very few had no significant diagnosis made. It is arguable, therefore, that these admissions were warranted; from a commissioner’s point of view and from a medico-legal point of view, there can be little to fault the decision to admit.

The GP’s risk though might well lie amongst those patients not admitted. The differential diagnosis of acute abdominal pain is wide. In the Brekke and Eilerston study, over 10 other conditions were diagnosed, the commonest being urinary tract infection and gastroenteritis. The UK Map of Medicine suggests 13 conditions applicable to either sex, with two additional suggestions for men, three for children and no less than five additional suggestions for females of child bearing age. Many of these conditions require significant diagnostics to confirm or even make a diagnosis. This is not the stuff of general practice. What is required is the ability to select those patients who require admission, or at least an opinion, thereby ensuring that the patient is safe and, in the process, minimising medico-legal risk.

History and examination

As always a good history, appropriately recorded, is essential. This may give the likely diagnosis or at least suggest the probable need for admission or not. Severe pain and/or marked systemic features can be very helpful, but elicitation and recording of important negatives can become crucial – especially if the decision is not to admit.

General questions with regard to the onset and progression of symptoms, particularly pain and any migration of pain, are a good starting point together with more generalised questions with regard to nausea, vomiting, bowel action and urinary symptoms. In women of child bearing age, a menstrual history, date of last period, sexual activity and contraceptive history are important. The presence or absence of vaginal discharge or intermenstrual bleeding may also be required. For children, the presence or absence of a history of sore throat should be elicited. For all a brief past history, especially of recurrent symptoms or significant co-morbidity is required.

An appropriate and appropriately recorded examination is also required – perhaps even more so where the decision is not to admit. The general appearance and demeanour of the patient tells me a lot. The presence of a pyrexia or tachycardia can be helpful, but are not invariably present in significant abdominal pain, especially in the early stages. Is there a foetor? The abdominal examination is probably most important. Is there tenderness and where? Is there guarding, a mass, an acutely tender gall bladder and (for me especially) is there rebound tenderness? Are the loins and hernial orifices clear?

With regard to intimate examination, cognisance of any working diagnosis together with the likelihood of the examination furthering the diagnosis and the availability of appropriate equipment and chaperone all need consideration. Where ectopic pregnancy is suspected, then vaginal examination should be avoided. However, a working diagnosis of pelvic inflammatory disease together with appropriate facilities including the option to take swabs (especially where the patient’s general condition is not suggestive of the need for admission), suggests examination and prompt treatment in the community may be beneficial. Where a decision to admit has already been made, then rectal examination need not be performed. If there is doubt with regard to admission, then a rectal examination may be helpful, perhaps revealing significant tenderness, or unexpected blood, mucus or pus. When not minded to admit then testicular examination should be performed, especially in young men.

Urinalysis, especially using sticks and including leucocytes and nitrite, can be helpful but the timeliness of other examinations, often with poor sensitivity and specificity (for example full blood count and CRP), makes their use in the community for urgent cases less helpful and may delay an appropriate admission.

To admit or not admit

The decision to admit a patient with acute abdominal pain, and to even form a definitive diagnosis, can be a very easy one. Classical appendicitis or an acute perforated duodenal ulcer may be obvious, but the presentation of retro-caecal appendicitis or a diverticular perforation can be difficult. The safe handling of the patient is paramount and on occasion this will lead to admission where little is subsequently found.

The decision not to admit requires consideration of appropriate follow-up and possibly laboratory or other tests. Safety netting is crucial. A patient with acute abdominal pain not admitted but ill enough to require reassessment the same day (unless a child) probably does need admission – not least because a surgeon assessing such patients regularly will have greater current experience than most GPs.

Today most hospitals have an acute ultrasound service for both general surgical and gynaecological purposes. Whether patients with a classical presentation require such investigation is a different story for commissioners. But as a GP, I will continue to be guided by my patient’s history, examination findings and on occasion intuition. This is likely to provide the best care for my patients and allow me to sleep at night – but if something does turn out wrong I will have a defensible stance.

Dr Jonathan Berry is a general practitioner in Trafford and a healthcare management consultant

*Brekke M, Eilertsen R K. Acute abdominal pain in general practice: tentative diagnoses and handling. A descriptive study. Scand J Prim Health Care. 2009; 27(3): 137–140