Quick blood draw sorts tonsillitis from mono

White cell ratio more accurate than costlier spot test. Frees up valuable lab time

By Gloria Gaspard

Not sure if a patient has bacterial tonsillitis or mononucleosis? All you need is a blood sample and a calculator, say a team of British researchers. The ear, nose and throat specialists from St Georges Hospital, London, have established that the ratio of lymphocytes to total white cell count (L/WCC) is as diagnostically accurate — if not more so — as the current gold standard and could easily be used by physicians to guide treatment decisions, they write in a paper published in the January issue of the Archives of Otolaryngology Head and Neck Surgery.

Bacterial tonsillitis and mononucleosis, also known as glandular fever, can be hard to tell apart. The symptoms — sore throat, fever, difficulty swallowing, redness of the throat and tonsils and white plaques on the tonsils — are similar and making the distinction is important in determining treatment (see flowchart, right). "Bacterial tonsillitis you treat with antibiotics, so you don't want to do that for nothing," says Dr Saul Frenkiel, president of the Canadian Society of Otolaryngology. "It's also important to know that someone has mono because of other implications. It's a longer recovery, for one, and the patient needs to know that. They also have to have liver function tests and be placed on certain restrictions to avoid rupture of the spleen." The spleen is enlarged in patients with mono.

THE GREAT DIVIDE
In their study Dr Dennis M Wolf and colleagues analysed samples from 120 patients with mono and 100 patients with bacterial tonsillitis retrospectively. All subjects received a mononucleosis spot test — which checks for antibodies specific to the Epstein Barr virus — and had their L/WCC ratio calculated.

Patients in the tonsillitis group had significantly higher total white blood cell counts, while the number of lymphocytes was significantly increased in the glandular fever group. The number of neutrophils was also elevated in patients with tonsillitis compared to those with glandular fever.

Most important, however, was the realization that the mean L/WCC ratio was very different in the two groups: patients with mono had a mean L/WCC ratio of 0.54, compared to 0.10 for those with bacterial tonsillitis.

On the basis of these results, the group determined that a L/WCC ratio of 0.35 would give their diagnostic test a specificity of 100% and a sensitivity of 90% for detecting mono — a better record than the mononucleosis spot test itself, the authors write. A 1996 study found that six commercially available rapid test kits had sensitivities ranging from 70-92% and specificities of 96-100%.

False diagnoses from the spot test are relatively rare, but can have tragic consequences. A group of Dutch researchers recently documented the case of an 18-year-old man who rapidly progressed to multiorgan failure after a positive mononucleosis spot test result. False positives have also been documented in patients with leukemia or lymphoma, rubella, hepatitis and systemic lupus erythematosus.

At the very least, the authors conclude, "the L/WCC ratio should be used as an indicator to decide whether mononucleosis spot tests are required."

NOT A REPLACEMENT
Dr Frenkiel says that while the study is an interesting academic exercise, he isn't sure physicians will be ready to give up on the mono spot test, even if the numbers do look a little better. "What if the ratio was 30 and not 35?" he asks. "You're going to do the mono test." And he doesn't believe it would be much of a time saver, either. "Whether you draw blood for lymphocyte and white blood cell counts or for a mononucleosis spot test, it's still the same specimen," he explains. But cost must be taken into consideration as well, the authors point out: "[This test] would prevent unnecessary requests for random and expensive mononucleosis spot tests and facilitate the use of appropriate treatment regimes resulting in possible shorter hospital stays."

By their own admission, further prospective studies are needed to confirm the reliability of their method. But they're confident that the L/WCC ratio can be relied upon to determine which patients should undergo further testing "to avoid unnecessary stress on limited laboratory resources."