Liver Involvement with Mycoplasma pneumoniae Community-Acquired Pneumonia

I read with interest the report by Grüllich et al. reporting cholestatic hepatitis in an adult male with community-acquired pneumonia (CAP) (5). I believe there are several problems with this case.

The authors base their assumption on the temporal relationship between Mycoplasma pneumoniae CAP and cholestatic hepatitis. As the authors point out, this is exceedingly rare, if it occurs at all in adults. The patient described is a 35-year-old man with “high fever,” which is not characteristic of M. pneumoniae infection, in which fevers are usually equal to or less than 102°F. The infiltrates were described as being dense in the right lower lobe, which is also uncharacteristic of Mycoplasma, which usually has infiltrates that are not dense. Also disconcerting is the presence of bilateral pleural effusions in a patient with a unilateral pulmonary process, and ascites were described, which are again not characteristic of M. pneumoniae CAP. The authors' diagnosis of M. pneumoniae was made on a serological basis. The authors further state that the diagnosis was supported by the fact that the patient initially improved after receiving clarithromycin therapy. Many features described are inconsistent with the diagnosis of M. pneumoniae CAP, so questioning the diagnosis is reasonable.

M. pneumoniae CAP is characterized by the absence of liver involvement, temperatures equal to or less than 102°F, the absence of relative bradycardia, and a persistent nonproductive cough that lasts for days to weeks. Extrapulmonary manifestations include nonexudative pharyngitis, meningoencephalitis, erythema multiforme, myocardial involvement, or watery diarrhea. The most common clinical presentation is that of CAP with mild nonexudative pharyngitis, dry and unproductive cough, and watery diarrhea. The authors do not comment on diarrhea or the persistence of dry cough. They also do not describe the temperature or the relationship of the pulse to the temperature, which would be helpful diagnostically. Another critical omission is the absence of cold agglutinin titers in the case reported. Approximately 75% of patients with M. pneumoniae CAP have elevations of cold agglutinin titers that are often in the range equal to or greater than 1:64. This is not described in the case. The chest X-ray findings typical of M. pneumoniae include unilateral, patchy, but not dense infiltrates, plus or minus a small pleural effusion on the involved side. Bilateral effusions and ascites have not been ascribed to M. pneumoniae (2, 3, 6).

In this case, the persistence of alanine aminotransferase (ALT) elevations long after clinical resolution of the CAP is most likely due to the clarithromycin and not from an immune basis. Clarithromycin's ability to cause cholestatic hepatitis is well known (1, 4).

In summary, the authors conclude that cholestatic hepatitis should be included in the differential diagnosis of M. pneumoniae CAP. I strongly disagree, as this would lead clinicians away from the correct diagnosis more often than not. The causes of CAP that commonly present with liver involvement are Legionnaires' disease, psittacosis, and Q fever. Since no epidemiological information was included with the case, there is no reason to suspect psittacosis or Q fever in the differential diagnosis. However, Legionnaires' disease is not uncommon, even in 35-year-old adults, and hepatic involvement with Legionnaires' disease is a characteristic feature. The presence or absence of hepatic involvement is one of the key points in differentiating Legionnaires' disease from M. pneumoniae CAP. High fevers with a pulse-temperature deficit support the diagnosis of Legionnaires' disease and argue against the diagnosis of M. pneumoniae CAP. Similarly, the dense infiltrates on chest X-ray suggest Legionnaires' disease rather than M. pneumoniae. Serologies for Legionella, Q fever, and psittacosis were not included in the case report. The patient described also had scleral icterus and mild jaundice. Again, this is not common in Legionnaires' disease but is rare with other causes of atypical CAP, e.g. M. pneumoniae.

Coinfections are exceedingly rare in CAP, so Legionella plus Mycoplasma coinfection does not provide a credible explanation in this case either. The magnitude of the elevation of the immunoglobulin M (IgM) and IgG enzyme immunoassay Mycoplasma titers suggests that the serological findings are best explained by a recent antecedent M. pneumoniae infection. They were also very high for an infection that began 2 weeks previously. This would explain why IgM and IgG titers were still elevated, as M. pneumoniae IgM titers may remain elevated for 6 to 12 months after the acute event (2, 3, 6).

In a patient with CAP, hepatic involvement should suggest Legionella infection, psittacosis, or Q fever rather than M. pneumoniae infection. None of the clinical findings in this case are typical of M. pneumoniae but are consistent with atypical CAP infection, particularly Legionnaires' disease. Therefore, the presence of liver involvement in a patient with CAP should argue strongly against the diagnosis of M. pneumoniae CAP in adults.