Article Text

Download PDFPDF

C-reactive protein: a valuable acute investigation. A case of pneumococcal meningitis presenting as ankle pain
  1. J S Huntley,
  2. M B Kelly
  1. Musculoskeletal Research Unit, University of Edinburgh, Teviot Place, Edinburgh EH8 9AG
  1. Correspondence to:
 Dr J S Huntley
 Musculoskeletal Research Unit, University of Edinburgh, Teviot Place, Edinburgh, EH8 9AG;


A case is presented in which the decision to admit and treat an adult with musculoskeletal pain and pyrexia was based on her markedly raised c-reactive protein (CRP). At the time of admission she was apyrexial and the CRP was the only haematological investigation that was out of the normal range. She subsequently became precipitously septic with pneumococcal bacteraemia and meningitis. The CRP is an important investigation for emergency departments.

  • CRP, C-reactive protein
  • FBC, full blood count
  • GCS, Glasgow Coma Scale
  • WBC, white blood count
  • C-reactive protein
  • CRP
  • bacteraemia
  • meningitis
  • soft tissue infection

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


A 53 year old woman presented to the emergency department with a one day history of “flu-like” malaise, intermittent fever, and atraumatic left ankle pain, which, though poorly localised, was of increasing severity. Her medical history included episodic migraine and hyperthyroidism treated with radioiodine rendering her hypothyroid (well controlled with supplementation). On examination she was pyrexial (38.5°C) but normocardic. She had a full range of pain-free movement at the ankle joint, which was not swollen, tender, or warm, and showed no evidence of an effusion. However, she was only able to weight bear with discomfort. Ankle radiographs showed no abnormality. Blood tests taken were full blood count (FBC), urate, and C-reactive protein (CRP). She was allowed home with a diagnosis of viral infection and “arthralgia”, and was given an appointment for review with results (CRP processing was not available overnight) at the fracture clinic the subsequent morning.

At review, the CRP of 123 mg.l-1 and leucocyte count of 11.4 (neutrophils 9.8) × 109.l-1 was her only abnormal results. Septic arthritis was considered but clinically there was no effusion and an aspiration was dry. She had a full active and passive range of movement at the ankle, with discomfort only when stressed at the extreme of dorsiflexion. There was no tenderness on bony palpation. A reactive arthritis was not thought likely because of the contemporaneous nature of her general symptoms and musculoskeletal pain. She was tender over the tendons of tibialis anterior and extensor hallucis longus. A provisional diagnosis of inflammatory/infective tendonitis/synovitis was made. Though she was apyrexial and only slightly tachycardic (PR 96 bpm; BP 120/75 mm Hg), because of her markedly raised CRP she was admitted for further investigations, close observation, and antibiotic treatment. At this time, repeated bloods showed white blood count (WBC 9.9 × 109.l-1), glucose, urea, electrolytes, and urate within reference ranges. Urinalysis was also negative.

After blood cultures, she was started on “best guess” intravenous antibiotics (flucloxacillin 1 g and penicillin G 1.2 g). Two hours later, she became septic with two rigors, a spiking pyrexia (39.5°C), and tachycardia (PR 104 bpm; BP initially 110/70 mm Hg). She also complained of a throbbing headache with photophobia. Kernig’s sign was positive. She was not tachypnoeic, had no adverse respiratory signs, and oxygen saturations were 99% on air. Her Glasgow Coma Scale (GCS) remained 15 and there were no focal neurological signs and no cutaneous rash. Plantar reflexes were bilaterally down-going and fundoscopy was normal. A coagulation screen was normal.

Her blood pressure then dropped to 90/60 mm Hg (15 minutes after previous reading), but responded to fluid resuscitation. A chest radiograph and urgent head CT (result available at 1 hour post onset of sepsis) showed no abnormality. A subsequent lumbar puncture of clear colourless fluid showed red blood count 200 mm-3, WBC 312 mm-3 (90% neutrophils), glucose 2.7 mmol.l-1, protein 2125 mg.l-1, and was negative for meningococcal, Haemophilus influenzae and pneumococcal antigens (delayed culture negative). She was transferred to the care of the infectious diseases team.

Blood cultures yielded chains of Gram positive cocci and were pneumococcal antigen-positive at 15 hours. She was treated with 14 days of intravenous cefotaxime (2 g qds) followed by oral amoxicillin for one week. A throat swab taken previously showed “normal commensals only”. At discharge (day 20 post admission) her CRP was <10 mg.l-1. She made a complete recovery. Six weeks after discharge, a repeat ankle film showed no abnormality.


Streptococcus pneumoniae is an important pathogen, with bacteraemia carrying a mortality of 16–24%.1,2 A previous report in Scotland1 identified the primary site of pneumococcal sepsis being the lungs in 76%, the meninges in 6%, joints in 2%, and soft tissues only 1%. However, in 15% no site was identified and most patients (62%) had recognisable predisposing factors.1 A high rate of second site infection—for example, meningitis, endocarditis—has been documented for pneumococcal arthritis.3 We think it most likely that our patient—who did not have any predisposing factors—had pneumococcal bacteraemia with meningitis and a soft tissue focus. It is impossible to establish the primacy of these sites. An analysis of adult cases of bacterial meningitis in Munich showed a complication rate of 50%, and a mortality of 18.6%, that rose to 33.3% (10/30) for pneumococcal cases.4

CRP, so called because of its reactivity with phosphocholine residues of C-polysaccharide of Streptococcus pneumoniae,5 is a pentameric acute phase protein synthesised by hepatocytes.6,7 Expression is predominantly regulated at the level of transcription, with interleukin-6 being the predominant inducer. In health, levels rarely rise above 10 mg.l-1. It is generally ascribed a limited role in the diagnosis of adult disease in the emergency department. The review of Clyne and Olshaker (1999)8 concludes that “A single CRP value should not factor into the decision to treat...”. In our case, however, the raised CRP indicated a severe underlying condition, even though the condition itself had not been diagnosed. At the time of admission, she was apyrexial with a FBC conforming to the normal range. When she became precipitously septic, it was fortuitous that she was already a hospital inpatient. She was treated rapidly and made a complete recovery from a condition associated with high mortality and morbidity.

In summary, we have reported a case of pneumococcal bacteraemia and meningitis presenting as ankle pain with fever, in which the CRP was the sole investigation mandating her admission. There are two lessons: (1) localised musculoskeletal pain with pyrexia should prompt thought of a soft tissue infection if septic arthritis and osteomyelitis are held to be unlikely, and (2) in cases of diagnostic uncertainty, the CRP can be a useful one off investigation in the emergency department. We therefore believe that acute hospitals should have access to CRP processing and results services 24 hours a day, seven days a week.


We gratefully acknowledge the encouragement and advice of Mr IGC Weir, Consultant Orthopaedic Surgeon (Queen Margaret Hospital, Whitefield Road, Fife), in writing this case report concerning his patient.



  • Competing interests: none declared

Linked Articles

  • Primary Survey
    Geoff Hughes