What is the evidence that "HIV" causes PCP? 29 November 2004
Previous Rapid Response Next Rapid Response Top
Eleni Papadopulos-Eleopulos,
Department of Medical Physics, Royal Perth Hospital, Western Australia, 6001,
Valendar F Turner, John Papadimitriou, Barry Page, David Causer, Helman Alfonso, Sam Mhlongo, Todd Miller, Christian Fiala

Send response to journal:
Re: What is the evidence that "HIV" causes PCP?

Where is the evidence that "HIV" causes PCP?

In his rapid response: "Re: PCP and “HIV"”, 9 November, Peter Flegg wrote: "It appears as though the Perth Group, having spent the last 20 years trying to alter the rules of virology, is now trying to drive a coach and horse through the fields of orthodox microbiology, clinical medicine and epidemiology. Pneumocystis typically occurs in a setting of immunodeficiency. The predominant cause of this is HIV infection. Attempts by the Perth Group to pretend there has not been a dramatic epidemic of PCP since the start of the HIV epidemic are doomed to failure".

We have never claimed that in the last 25 years there was not an increase in PCP. Our questions were:

1. Is the rarity of PCP reporting before the 1980s due to its very low frequency or is it because before the 1980s (a) reporting of PCP was always based on a definite diagnosis; (b) it was generally accepted that definite diagnoses can be obtained only by performing an open lung biopsy; (c) open lung biopsy is an invasive procedure. Because of this it was rarely used for the differential diagnosis of lung disease. In our rapid response: "PCP and "HIV", 5 November, we wrote: "Would Peter Flegg please tell us how often he was requesting open lung biopsies for the differential diagnosis of lung disease in his drug abusing patients, or any other patient, in "the pre-HIV era" or even today?" Peter Flegg did not respond.

2. Is it possible that the very high frequency of PCP reported in the last 24 years in gay men, haemophiliacs and drug users is, at least in part, due to the use of unreliable "definitive" diagnosis?

A specific diagnosis requires the demonstration of P. carinii in the affected lung tissue. A standard surgical open-lung biopsy provides the most satisfactory specimen for diagnostic studies…Transbronchial biopsy or endobronchial brush-biopsy techniques have also been used to obtain tissue specimens, but these methods are not quite as dependable as the open-lung biopsy".[1]"…one might expect to find P. carinii in the fluid from bronchoalveolar lavage of about 40% of patients with AIDS who present with symptomatic pneumonia caused by other organisms".[2] Yet according to the "HIV" experts "…bronchoalveolar lavage" is an "accurate" method for the definite diagnosis of PCP.[3] According to PCP and "HIV" experts "Although the clinical features of P. carinii pneumonitis are quite constant, similar manifestations occur with other pneumonitides in the compromised host”;[1]; "clinical standards alone" are "inaccurate criteria" for the diagnosis of PCP". According to the CDC, the following can be used only for a "presumptive" diagnosis of PCP:

(a) A history of dyspnoea on exertion or unproductive cough of recent onset (within the past three months); and

(b) Radiological evidence of diffuse bilateral interstitial infiltrates or gallium-scan evidence of diffuse bilateral pulmonary disease; and

(c) Arterial blood-gas analysis showing an arterial partial pressure of oxygen of less than 70 mmHg (9.33 kPa) or a low (less than 80% of predicted values) respiratory diffusing capacity or an increase in the alveolar-arterial oxygen tension gradient; and

(d) No evidence of a bacterial pneumonia.[4]

We asked Peter Flegg what method he is using to diagnose PCP in his patients. He did not respond. He only states that PCP "is a very characteristic clinical and radiological type of pneumonia". The reference which Peter Flegg gave to show that in Africa PCP is actually "far commoner than many realise (or at least those who fail to keep abreast of developments in the field)", suggest problems with diagnosis. Fisk et al wrote: "Autopsy reports also showed that disseminated infection with Mycobacterium avium complex and PCP occurred less frequently among HIV-infected patients in Africa than among such patients in developed countries. Other studies showed that tuberculosis was the diagnosis most frequently given to selected HIV-infected patients who had abnormal chest radiography findings that were consistent with PCP. Citing these reports, the authors of texts on tropical medicine concluded that tuberculosis was the major HIV-associated illness in the developing world and that PCP was uncommon…During the first decade of the AIDS pandemic, PCP rarely occurred in African adults. More recent reports have noted that PCP comprises a significantly greater percentage of cases of pneumonia that it did in the past". Fisk et al added: "It remains speculative whether the trend toward an increase in the proportion of African cases, as illustrated in figure 1, denotes a true increase in the prevalence of PCP or whether the early reports underestimated the actual prevalence".

Given that: "Throughout the developing world, the rate of coinfection with Mycobacterium tuberculosis and PCP is high, ranging from 25% to 80%" and the more "sensitive" tests used for the "definite" diagnosis more recently are non-specific, the increase may be an artefact.

3. What is the cause of the increased incidence of PCP in the last 25 years?

In 1977 one PCP expert wrote: "…the impaired cell-mediated immunity associated with severe protein malnutrition unite these individuals into a population susceptible to P. carinii pneumonitis as well as other opportunistic infections”.[1] Many Africans, gay men and drug abusers are at risk of malnutrition. In the Fisk paper, it is interesting that in one study PCP accounted for pneumonia in only 5% of children, 40% of whom were not infected with "HIV". "One study of patients in Cote d'Ivoire who died of any cause during 1989-1991 and another study of patients in the same country who died of pneumonia in 1989 showed that the prevalence of PCP was 3% and 13% respectively. In the latter series, one-third of the patients were not infected with HIV". In a 1964 publication on PCP one reads: "Children and adults with agammaglobulinemia, lymphomas, or leukaemias, and those receiving cytotoxic drugs, antibiotics, or corticosteroids may also be affected…The diagnosis of Pneumocystis pneumonia is usually made post mortem".[5] Since 1964 there has been an increase in the prescription of antibiotics and steroids, in both the general population as well as the AIDS risk groups. In addition, gay men, drug users and haemophiliacs are exposed to immunosuppressive agents (semen, drugs).

Peter Flegg claimed that the patients with PCP described by Jacob et al had many diseases and thus "were not as immunocompetent as the Perth Group would have us believe".

In our rapid response PCP and "HIV", 5 November, we asked Peter Flegg: "If diabetes, CCF, asthma, alcohol abuse, cardiac valve disease, lower immunocompetence, which in turn predisposes to the development of PCP, then given the fact that these type of patients: (a) have been present in the New York Hospital-Cornell Medical Centre long before the 5 cases were reported, as well as since then, why was PCP reported only in this small period of 3 months; (b) are present in every major hospital around the world, why "PCP in the absence of HIV immunodeficiency is exceedingly rare?" Could it be because patients other than those suspected of having AIDS are not tested? Is it possible that the reported high frequency of PCP in the Jacobs et al study and in many, if not all AIDS patients, is not real but the result of a non specific test?"

Peter Flegg did not respond.

In conclusion, there are many reasons for the rarity of PCP reports before the AIDS era and for their increase since 1980. Peter Flegg, on the one hand claims that diabetes, CCF, asthma, alcohol abuse, cardiac valve disease, lower immunocompetence, which in turn predisposes to the development of PCP, and on the other hand the high incidence of the disease in the AIDS risk groups is not malnutrition, semen, factor VIII or drugs, but "HIV". The only evidence he gives in support of his claim that "The predominant cause of this [PCP] is HIV infection" is a correlation between a positive "HIV" antibody test and PCP.

In our rapid response: "PCP and "HIV"", 5 November, we wrote: "However, as we have stated many times before (something which the "HIV" experts including Peter Flegg chose to ignore) the question is not if a positive "HIV" antibody test is an indicator for present or future development of AIDS, but does it prove infection with "HIV". The absolutely necessary but not sufficient condition for any scientist worth his salt to claim or accept that "HIV" is the cause of AIDS, is evidence that a positive antibody test is proof of infection. In this debate, we have drawn the attention to the "HIV" participants, including Peter Flegg, that according to the test kit manufacturers and, most importantly, to some of the best known experts in "HIV" testing, at present no evidence exists that a positive test means infection. For some unknown reason all of them, including Peter Flegg, have ignored this fact. We asked them to provide evidence that the "HIV" antibody tests are specific. So far no such evidence has been forthcoming".

In his rapid response: 'Re: PCP and "HIV"", Peter Flegg once again ignored these facts.

So we repeat: "If Peter Flegg and his colleagues fail to provide such evidence, then we must conclude that their argument for a causal relationship between "HIV" and AIDS (PCP), based on a correlation between a positive antibody test and AIDS, collapses and any further debate is superfluous" and scientifically baseless.


1. Hughes WT. Pneumocystis carinii pneumonia. N Engl J Med 1977;297:1381-3.

2. Hughes WT. Pneumocystis carinii pneumonitis. N Engl J Med 1987;317:1021-1023.

3. Fisk DT, Meshnick S, Kazanjian PH. Pneumocystis carinii pneumonia in patients in the developing world who have acquired immunodeficiency syndrome. Clin Infect Dis 2003;36:70-8.

4. Whyte BM, Cooper DA. The surveillance definition of the acquired immunodeficiency syndrome and the clinical classification of infection with the human immunodeficiency virus type 1. Med J Aust 1988;149:368-73.

5. Rifkind D, Starzl TE, Marchioro TL, Waddell WR, Rowlands DT, Hill RB. Transplantation pneumonia. J Am Med Assoc 1964;189:114-118.

Competing interests: None declared