Research into Toxic Shock Syndrome

Please read the Legal Page before continuing.

Most of the microbiological research into what triggers TSS has been carried out in the USA, and generally sponsored by tampon manufacturers.

Of particular interest was a review of other research papers undertaken by Jeffrey Parsonnet MD, Associate Professor of Medicine and of Microbiology at Dartmouth-Hitchcock Medical Center, New Hampshire. It is believed that the purpose of this paper was to justify the Tampax claim that a tampon could be used safely for up to 8 hours. Tampax took this research a stage further and claimed "overnight - up to 8 hours". However, the Parsonnet paper highlighted 2 more significant risk factors, viz. CONTINUOUS USE of tampons during a period, and that higher ABSORBENCY increases the risk of Toxic Shock Syndrome.

Parsonnet's paper is entitled "Length of Wear of Tampons and Risk of Toxic Shock Syndrome".

He writes,

"I have reviewed the relationship between length of wear of tampons and the risk of developing TSS. Several epidemiological studies have collected and analyzed data related to length of wear. The Tri-State Study (of Osterholm, Davis and Gibson 1982), revealed no significant association between length of wear and risk, although there was a trend towards increased risk when tampons were worn for 13 hours or longer. Studies performed by the (US) Centres for Disease Control (CDC) have also failed to show any association between length of wear and risk. (These studies were undertaken by Shands in 1980, Schlech in 1982, and Reingold in 1989). However, the Reingold paper showed that patients with TSS tended to use more tampons per day than their matched controls. This suggests that the average duration of wear of each tampon may have been shorter, rather than longer, among cases than among controls. TSS cases were more likely than controls to have used tampons CONTINUOUSLY on at least one day of their period, and the mean maximum time that a single tampon was left in place was slightly longer among patients than controls (i.e. 7.8 hours as opposed to 6.6 hours). Once absorbency and continuous use were taken into account, however, maximum length of use was NOT an independent risk factor for development of the disease. In summary I believe that there are no epidemiological, clinical or laboratory data to suggest that a recommended length of tampon wear of 6 to 8 hours would increase the risk of a woman developing TSS."

AKTA wondered whether the use of the word "woman" in the conclusion included "girl". Other research has concluded that another factor in developing TSS is that the body has low immunity to the TSS toxins, and girl's immunity is much lower than women's immunity. Therefore, we conclude that tampons should be used for no more than 6 hours, and that a sanitary towel should be used at night to break CONTINUOUS USE.


The (US) Journal of Adolescent Health Care of July 1986 featured an article written by Dr. Lawrence D'Angelo. He reported that teenagers continue to comprise a disproportionate number of recorded cases of Toxic Shock Syndrome, despite the fact that the majority of teens do not use tampons. There are several possible explanations for this increased risk:

"Whatever the explanation, if the overall number of TSS cases is decreasing, why should we be concerned? For the past several years, the manufacturers of tampons have increased their marketing and advertising efforts to recruit young teenagers to use their products. Several manufacturers have accommodated the younger teen by making thinner tampons that are easier to insert. Most of these are of low absorbency and are felt to pose a minimum risk. However, if young teens use a range of different size tampons regularly, it may well significantly increase the teens risk of acquiring TSS."
AKTA is concerned that thinner tampons contain more of the absorbent man-made fibre (rayon). This puts the younger teenager at even greater risk of developing TSS. Since it is recognised that tampon fibre ABSORBENCY is the most critical of the Risk Factors, it is useful to report here of a summary of a research paper produced by independent researchers Tierno and Hanna of New York University Medical Center. (Independent means that the research is not sponsored by the tampon manufacturers).


Philip M Tierno & Bruce A Hanna - New York University Medical Center

"Historically, research into TSS has suggested that TSS-T1 is stimulated by various ecological factors, (e.g. presence of oxygen or carbon dioxide, or lack of magnesium), many of which have been shown to stimulate production of TSS-T1. Here we report on the propensity of different fibres used in tampons to amplify TSS-T1 production in TSS associated strains of Staphylococcus aureus.
Two hundred and twenty intravaginal devices were used. It is concluded that certain materials provide a more absorbent fibre where nutrients are efficiently drawn in between the fibres and thereby creating an ideal physiochemical environment for the amplification of TSS-T1 and other toxins. The greatest stimulation of TSS-T1 was observed with (in decreasing order): polyester and carboxymethyl cellulose (CMC), polyacrylates, viscose rayon, gelatin foam, polyurethane, and cotton."

AKTA draws the conclusion that the more absorbent the fibre, then the higher risk of developing Toxic Shock Syndrome. Fortunately tampon manufacturers no longer use polyester and carboxymethyl cellulose. However, most tampons are a combination of rayon and cotton. Cotton is the least absorbent material, but appears to be the safest. There are all cotton tampons available, but not in the super absorbent size because they would have to be quite large. Just to refer back to teenage tampons, it can be seen that the slim tampons obviously contain more rayon.


Dr Patrick Schlievert, who has carried out research on behalf of Proctor and Gamble (P&G made the infamous "Rely" brand tampon which killed most of the 40 women in the USA in 1980) and Dr R. Syverson who has carried out research for Kimberly-Clark Corporation, claimed that they have been unable to replicate the findings of Tierno and Hanna. A close inspection of their methodology reveals significant differences in technique. The debate continues!


Studies in rabbits indicate that small amounts of TSS-T1 appear to trigger the onset of TSS when introduced into the vagina, than when it enters the body in other ways. A study carried out by Dr. Marian Melish of the University of Hawaii, showed that onset of TSS occurred more rapidly than with subcutaneous or intravenous injections, which also caused less damage.

In these laboratory studies, TSS-T1 spread from the vagina into the blood and urine within 30 minutes of being introduced, and caused severe, rapidly progressive TSS at doses as low as 18.5 microgram. This dose is much lower than that capable of producing TSS from a subcutaneous focus. The vaginal mucosa seems to be a special site for the toxin.


Dr JP Arbuthnott, Department of Microbiology, Moyne Institute, Trinity College, Dublin, Ireland.

In this 1987 research paper, Dr Arbuthnott reports that in the USA up to 1985 there were around 3,000 reported cases of TSS, with an overall fatality rate of 5.6% and an incidence of around 3 cases per 100,000. An unusual geographical distribution of TSS has been noted, with more than 40% of reported cases occurring in only five states (Wisconsin, Minnesota, Colorado, Utah and California). Outside the United States the disease has been less common: in Britain for instance there were 99 reported cases up to 30th June 1984 with an incidence rate of 0.06 per 100,000.

Risk factors in TSS: the role of tampons.

The initial high association of TSS with menstruation led to investigation of the possible role of catamenial products as risk factors. The use of tampons emerged as a significant risk factor in case controlled studies. In one case, a 15 year old, previously healthy girl in the third day of her menstrual period, suddenly developed shaking, chills, fever, severe muscle pain and vomiting. She developed diarrhoea and over the next two days her temperature rose to 40 degrees C. Within hours of admission to hospital, her blood pressure fell dramatically and she became disorientated and combative. There was mucous membrane involvement as indicated by conjunctivitis and vaginal discharge. Large amounts of crystalloid and colloid fluids were given intravenously to maintain the systolic blood pressure at 90 mm Hg. The patient remained acutely ill in Intensive Care for 3-4 days. Desquamation (severe peeling of skin) began on the toes and soles of the feet nine days after onset and then affected the hands. Cultures of blood, spinal fluid and urine were negative, while throat and stool cultures were normal. However cervical cultures grew Staphylococcus aureus and E coli.

The Center for Disease Control concluded that while tampons were not the only factor in pathogenesis, the risk of TSS could be eliminated by non-use of tampons and would be much reduced by intermittent, rather than continuous use.

One of the States mentioned above (Colorado) features in a more recent paper:


J.Todd, B.Kurtz, P.Combs, A.Todd and J.Anderson of the University of Colorado School of Medicine.

There are conflicting opinions regarding the current epidemiology of TSS. While studies using passive case ascertainment methods have suggested a dramatic decrease in the incidence of TSS since 1982, studies employing active case ascertainment methods have shown little or no decrease.

During the years 1990 - 1996, there were 35 cases of infection-related shock admitted to the hospital in Weld County. Of these 54% were related to Gram-positive organisms (40% due to Staphylococcal TSS and 14% caused by Streptococcus pyogenes). There was no significant difference for the incidence between the two time periods 1970 to 1982, and 1986 to 1996. During the latter time period, 79% of cases occurred in females of menstruating age (10 - 45). When cases from Larimer County were included, only 25% of cases coded specifically as TSS were passively reported to the State public health authorities. A greater proportion of cases of TSS satisfied the "possible" definition in the more recent era and there was a significantly lower length of stay (in hospital), suggesting that cases are being detected and treated earlier in the course of the illness, perhaps due to the widespread awareness of TSS among physicians and menstruating women. There were no deaths from TSS in any patients in this study.

The following conclusion can be reached about TSS in Colorado from 1970 to 1996. Gram-positive toxic shock comprises a substantial proportion of all infectious shock. TSS has not decreased significantly in incidence in at least one county in Colorado and menstrual cases still predominate. Cases appear to be recognised and treated earlier, however passive reporting to the State public health authority is incomplete and remains a poor estimate of overall incidence.

More research in another of those USA states mentioned by Arbuthnott:


A.Reingold, M.Manos and L.Hurley, School of Public Health, University of California, Berkeley, USA.

Staphylococcal TSS and its association with menstruation and tampon use received widespread attention in the US in 1980. Studies done in the early and mid 1980s demonstrated convincingly that the observed increase in reported cases of TSS, particularly menstrual TSS, was due, at least in part, to a true increase in incidence beginning in the late 1970s. However, these same studies did not demonstrate the consistent and marked decline in reported cases after 1980 that was seen in passive surveillance systems. The extent to which more recent temporal trends in the incidence of TSS, seen in cases reported through passive surveillance systems, reflect trends in the number of diagnosed cases is unknown. The study reported here attempted to examine temporal trends in the rate of diagnosed TSS.

Computerised hospital discharge data for all members of a Northern California Private Healthcare scheme for the time period 1981 to 1996 were searched for all individuals with a hospital diagnosis of staphylococcal TSS, whether as the primary or secondary discharge diagnosis.

A total of 265 members were hospitalised 273 times (eight members were hospitalised on two occasions each) and had a discharge diagnosis of TSS. Of these, 55 were male and 210 were female. Ten (eight women and two men) died. The number of members hospitalised for TSS each year ranged from three in 1981 (when the discharge diagnosis code for Staphylococcal TSS was introduced part way through the year) to 26 in 1992.

Toxic Shock Syndrome became headline news in Australia in 1994 following the death of 13 year old Peta-Ann Devine, a Queensland schoolgirl, who died suddenly whilst competing in a school swimming gala. Her death prompted the research below.


S.Garland, P.Robinson, M.Peel, G.Skowronski and G.Rough, Microbiology and Infectious Diseases Department, Royal Women's Hospital, Carlton, Melbourne, Australia.

Toxic Shock Syndrome occurs as a result of bacterial exotoxin produced by Staphylococcus aureus and Streptococcus pyogenes which causes a similar condition known as Streptococcal TSS (STSS). Both menstrual and non-menstrual TSS occur. During the Australian summer of 1994 - 95, menstrual TSS was the subject of considerable public attention. It was decided that a wider survey which examined the Australian experience of TSS from 1990 to 1994 would be timely.

As there is no coding for TSS in Victoria, the incidence of TSS had to rely on by active retrospective case finding through a telephone survey for the year of 1994. Further investigation was by contacting all Intensive Care Physicians through the Australian and New Zealand Intensive Care Society (ANZICS), and requesting information on cases from 1990 to 1994 by a mailout survey.

The telephone survey resulted in identifying 12 cases, (8 female and 4 male, 10 adults and 2 children). The number of tampon related cases was 4, and the main sequelae was loss of digits in one woman.

The ANZICS mailout survey targeted 220 ICU physicians and 74 (34%) responded. Of these 59 ICU physicians had not seen a case of TSS during the past 5 years. However, details of 23 TSS cases were submitted by 15 ICU physicians. All occurred on the eastern seaboard states. No deaths from TSS were reported among the cases over the period surveyed.


We are grateful to Dr Richard Tokarz of Washington, USA, for this research paper.


Toxic Shock Syndrome in women is always initiated by nausea vomiting and diarrhea resulting in dehydration; That severe dehydration results in dilation of the cervix up to 2 CM and that rapid dilation results in lacerations along the cervical canal; That the lacerations are essential to provide a path for the staff bacteria to have access to the body; That the lacerations cause large quantities of prostaglandin that is carried into the body and results in additional nausea, vomiting and diarrhea; That additional dehydration results in further damage to body tissue, causing more prostaglandin to be generated; That large doses of prostaglandin will cause the blood to thicken and blood pressure to drop and explain most of the other symptoms of TSS all the way to death.


The following is as concise an explanation of TSS as I can develop. It also defines what I believe to be the "toxin" in TSS, and what to do to mitigate the illness. More importantly, it defines the hazards of manual cervical dilation and the role of dehydration in cervical dilation and damage.

Twenty six years ago, I made an important discovery relating to the female human body. I was employed by Battelle Pacific Northwest National Laboratory. We received a contract funded by the U. S. Agency for International Development. At that time the government was funding abortion and, for example, there were 300,000 abortions in Washington D.C. in one year. Many of those abortions were called menstrual regulation. These required very low levels of cervical dilation. The rest of the abortions require higher levels of cervical dilation, as high as 15mm. The main objective of the contract was to find a way to measure the force required to dilate (OPEN) the human cervix.


There was very little definitive information in the literature, but it was apparent that every practitioner believed that the cervix was a stretchy tissue and the only requirement to dilate it was to force it open. For abortion or any other surgical procedure, the practitioner utilized Pratt or Hegar dilators. These consisted of a series of metal or plastic rods, the smallest was 1mm in diameter and each subsequent rod was 1mm larger with the largest rod being 16mm. These were inserted sequentially into the cervix using counter traction (1mm first and on to the desired dilation). It was common knowledge among practitioners that there was a noticeable drop in the force required to accomplish the dilation between 9 and 11mm. This was called "melting away". During manual dilation it was also noticed that there was a large infusion of prostaglandin into the bloodstream at higher levels of dilation. This had a side effect of headache, nausea, vomiting and diarrhea. Some practitioners thought that prostaglandin may be the reason for the "melting away" of the resistance to dilation. Practitioners tried to inject prostaglandin into the cervix prior to dilation. But the negative side effects of headache nausea, vomiting and diarrhea, as well as, the fact that there seemed to be no effect on the resistance to dilation ended those experiments. As far as I know, dilation today is still being done with these same dilators.


In the course of my initial work, I received freshly removed uteruses with cervix attached. My efforts to measure the force required to dilate these organs was frustrating, until I observed that the cervical tissue flowed around the blades of the measurement device. From this observation, I became aware that dilation of the cervix from 0 to 2cm was not a function of stretchy tissue. At the same time as I was trying to dilate cervical tissue, I enlisted the assistance of a local obstetrician. I asked him to measure the outside diameter of the cervices of pregnant women from 6 weeks prior to their reaching 2cm dilation prior to the onset of labor. He later reported that there was virtually no change in the outside diameter of his patients crevices as the cervix dilated from 0 to 2cm.

From these observations, I fabricated a dilator which dilated the excised cervix by compressing the outside of the cervix. This experiment verified the theory that dilation between 0 and 2cm was solely a function of removing excess fluid from the cervix. The detailed information discovered was presented at the first world conference on cervical dilation in Montreal and again at a meeting of Planned Parenthood, in Denver, and published in Battelle document, BN SA 1200, Sept. 1980. From the combination of observations and testing, I concluded that cervical dilation from 0 to 2cm is solely a function of removal of excess fluid, both in natural dilation by pregnant women and under manual dilation or in dilation by whole body dehydration. The problem here is that natural dilation occurs over five to six weeks and there is time for the tissue in the cervical canal to grow and expand without trauma, in forcible dilation or rapid dilation as a result of dehydration, (less than 72 Hrs), this tissue cannot accommodate the necessary stretching and lacerations occur along the cervical canal. In manual dilation, in addition to the lacerations, the fluid which is force up into the uterine tissue develops a hydrostatic pressure which is relieved by rupturing into the uterine cavity, This is what has been called "melting away" of the resistance to the higher dilations. These conclusions were verified in 1980, when my neighbour died of Toxic Shock Syndrome. She was 39 and on her menstrual period with super absorbent tampons. They were moving everything themselves into the house next door. She developed nausea vomiting and diarrhea and reported to the emergency room that same evening. She was diagnosed with the flu, given a IV and two Tylenol and told to see her doctor in the morning. By the next morning she was much worse but was able to drive to her doctor's office (across from the hospital) around noon and fell out of her car unconscious in the parking lot. She was transported to the hospital and died around 5:30PM. I called my doctor and asked him to contact the physician performing the autopsy and ask him to measure the level of cervical dilation, because I believed she would be dilated 2cm.

At autopsy, she was found to be dilated near 2cm and with several lateral laceration along the cervical canal, All infected with staphylococcus aureus. As far as I know, this was the first time this dilation was observed, however, about a month later, Dr. Soderstrom was called in to consult on a patient who was in her fifties and severely dehydrated with something other than TSS. During examination she was found to be dilated 2cm. Dr. Soderstrom reported these findings to the CDC.

In 1995 another friend died of TSS. She was in her forties and was operating three businesses. She became ill and ignored the problem for 24 hours. When she presented at a local hospital, she exhibited all the symptoms of toxic shock including very low blood pressure. The attending physician must not have recognized these, because he operated and removed her appendix and one ovary. She died shortly after the surgery. At autopsy, the coroner reported that she had both Staff and Strep infections and was dilated 2cm.


The super absorbent tampons can have only two components which are associated with TSS. The drying capability could cause or contribute to cervical dilation and the tampon could shorten the path for the Staphylococcus bacteria to reach the lacerations in the cervical canal. But this observation requires that the cervix must be dilated above 6 or 7 mm to cause lacerations in the canal, before the infection can have access to the body. However, dilation of the cervix injects prostaglandin into blood stream due to the damage to the cervix, causing continued whole body dehydration and further dilation of the cervix, causing more prostaglandin and a spiraling downward of the dehydration. In addition, whole body dehydration damages tissue throughout the body and that damage results in more prostaglandin.

In 1980, I talked to the doctor said to be an expert in prostaglandin. My recollection is that he was practicing at a children's hospital in Chicago. Without mentioning TSS, I inquired about overdoses of prostaglandin in the human body. His reply was that there would be nausea, vomiting and diarrhea, redness of the hands and feet and low blood pressure, and in the end the blood pressure would likely go very high and death would result. From the information provided it should be apparent that the precursor to TSS is dehydration and the added infection from other sources is a secondary problem. It also should be apparent that dehydration results in the generation of prostaglandin and that results in a thickening of the blood. If allowed to go far enough, this will result in death because the heart cannot pump against the resistance.

The final argument I can present is a case reported from the U.K. The woman presented with full blown TSS, but with one extremely unusual symptom. All of her toes appeared frostbitten. Obviously they were not frost bitten, but they were black and had no blood circulation. She survived but her toes had to be amputated. It might be said that she had poor circulation in her toes to start with, but the only thing that could have cut off all circulation would be prostaglandin.

It is well known that the function of aspirin is to attack prostaglandin anywhere in the body (Sir John Vane, 1982) and to quickly convert prostaglandin from e2 to f2 and on to something further. This implies the proper treatment should be to administer antibiotics AND ASPRIN or nsaids. TSS still occurs in the U.S. and is more prevalent in some other countries. However, it is important to note that aspirin is contraindicated for children under 19, due to the possibility of Reyes Syndrome, But if they are facing death one should risk.

An important further consideration is that every time a woman suffers from severe dehydration, regardless of the cause, she will be dilated 2cm and her cervix will be lacerated. This makes her cervix a site for other possible infections. This should be considered in the treatment regimen and aspirin should be considered for anyone who is severely dehydrated and in jeopardy of dying.

One final thought, manual dilation of the cervix above 9mm damages the cervical canal and ruptures the wall of the uterus. This leads to scar tissue in the wall of the uterus and could place subsequent pregnancies in jeopardy.

Back to the Index