Autopsy of bacterial vaginosis: a physiological entity rather than a contagious disease

Sir, In 1892 the German gynecologist Albert Döderlein (1860–1941) described the vaginal secretion as consistent with two types: (1) normal secretion-stark acidity and rich bacillary flora; and (2) pathologic secretion-low acidity and rich mixed flora of bacilli and cocci. The reason for the pathologic secretion was initially given as ‘excesse in venere’, or impure coitus repeated at short intervals. Döderlein did not elaborate further on the pathologic secretion (1). In 1955, a new disease entity was described-Haemophilus vaginalis vaginitis, a newly defined specific infection previously classified ‘non-specific vaginitis’ (2). H. vaginalis (Hv) vaginitis was described as a sexually transmitted disease and the Hv bacterium as the causative organism. This concept derived from the observations that (1) inoculums consisting of Hv bacteria from culture and from vaginal material of infected patients, when deposited in healthy vaginas, frequently produced a symptomatic Hv vaginitis, characterized by a shift of Lactobacillus (Lb) flora to Hv flora; and (2) Hv was detected in 96% of urethral specimens from the spouses of wives with Hv vaginitis (2, 3). In 1971, I presented an alternative concept on the origin of Hv vaginitis based on the premise that the vagina is not a reservoir of microbes only, but also a copulatory organ (4, 5). The Papanicolaou smears taken before and after unprotected coitus revealed that the Lb morphotype vaginal flora predominant in the precoital smears was replaced by Hv dominant flora in postcoital smears. This phenomenon and the distinct appearance of the two morphotypes of flora provided me with a reason to suggest: (1) the vaginal bacterial flora can change rapidly and completely from the predominance of Lb to that of Hv dominance in response to sexual intercourse; (2) the change is caused by vaginal coital transudate and male ejaculate; (3) thus the postcoital Hv flora denotes a physiological postcoital condition; and (4) the Hv bacterium is a natural organism in the vagina and is a biological variant of vaginal bacilli rather than an infectious microbe. These assumptions are in line with the observation that the male ejaculate can neutralize the vaginal acidity in order to protect the ejaculated spermatozoa (6). The resultant vaginal content would further contribute to the protection of spermatozoa, other than through the lower acidity, by immobilization of Hv bacteria on the epithelial cells; in a direct smear recognized as ‘clue cells’; and by a paucity of leucocytes (7). It seems obvious that the postcoital Hv flora is another type of physiological flora in sexually active women of reproductive age (4, 5). However, the basic acidity of the vagina would be restored shortly after coitus once the effect of coital fluids is over (8). In case the restoration of the precoital physiology is delayed, it is likely that the mildly acid vaginal content populated by the feeble Hv bacteria would be unable to resist the overgrowth of Hv flora by invading anaerobes: a symptomatic form of Hv vaginitis develops that corresponds to the symptomatic bacterial vaginosis (BV), in fact a return to the abandoned non-specific vaginitis. The monobacterial form would respectively correspond to the asymptomatic form of BV. A recent editorial called for an improved approach to the prevention of BV in view of its high recurrences and symptomatic occurrences (9), and a recent report on various aspects of BV pointed to the numerous adverse health outcomes associated with BV, although ‘the reason for BV remains unclear’ (10). It seems obvious that the origin of BV as a contagious disease is insufficient in providing a reason for successful prevention and for explaining the results of the pertinent studies. In a 1974 study (11), I recorded that the inoculation experiments, presented as evidence of the contagious nature of Hv vaginitis (2, 3), seemed inadequate since they failed to record sexual intercourse in the study subjects. Thus, it is likely that the results of the inoculation experiments were, in fact, a physiological outcome of unprotected coitus (4, 5). It should be noted that the occurrences of BV in virginal females (12) and in women who have sex with women (13) could be a result of long-lasting psychogenic, eventually with somatogenic, sexual stimulation that could induce and maintain ‘vaginal lubrication’, i.e. transudation of vaginal fluid capable of neutralizing the basic vaginal acidity to the degree that would favor the emergence of Gardnerella vaginalis flora and hence BV (8). The reliability of the physiological concept of Hv vaginitis/BV may be tested, for example by its ability to respond to the questions that have emerged from studies based on the prevailing concept of H. vaginalis vaginitis/bacterial vaginosis. Perhaps the most pertinent question on BV concerns prophylaxis and treatment. In face of the physiological concept, the base for prophylaxis and treatment would be to allow sufficient time for the female organism to recover, both mentally and physically, from a previous sexual act by abstaining from further acts for some time and not to engage in this repeatedly and at short intervals; a practice known to be harmful in the 1890s (1). It may well be that the offensive odor associated with BV (2) actually would signal the need for the female organism to recover. In brief, the key proposal of this study is that bacterial vaginosis is by origin a physiological postcoital condition, not a contagious disease. This knowledge is likely to make BV a compact entity, free from inherent controversies, and this knowledge would profoundly facilitate the communication between patient and doctor, which would ultimately lead to improved care and better mutual satisfaction.