The Story of The Unfortunate Experiment

The Unfortunate ExperimentThe Unfortunate Experiment

The full story behind the inquiry into cervical cancer treatment

Sandra Coney

Published by Penguin Books (NZ) Ltd,
182-190 Wairau Road, Auckland 10, New Zealand
First published 1988
Copyright © Sandra Coney, 1988
All rights reserved
Typeset by Typocrafters Ltd, Auckland
Printed in Australia 

Chapter three: The Story

Writing 'The Unfortunate Experiment' took months of Phillida and me sitting down together at the computer whenever we were in the same city. It was a long story, some 13,000 words, condensing twenty years of events into a final eighteen pages of Metro. During the later cancer inquiry, our story would come under intense scrutiny and still hold up. There were very, very few errors in it. We learned a great deal more in the months after publication; it did not alter but only expanded the story as we had told it in Metro. In the following retelling, some of this material is included. This then is the story.

National Women's Hospital inherited responsibility for the treatment of cervical cancer from Ward 4 at Auckland Hospital. It took cases from the whole of the northern half of the North Island and by 1954 a specialised clinical team had been set up.

In the days before cytology, cervical cancer was often not detected until an advanced stage. The only treatment was by deep ray or radium, although in the early fifties surgery was introduced at the hospital to follow irradiation of the pelvis.

With cytology, it was argued that the earlier 'precursor' stages would be detected, thus enabling the disease to be treated before it reached the invasive stage. It was thought that a smear from the cervix might reveal cell changes in their earlier stages.

The head of National Women's Hospital, Harvey Carey, knew that at nearby Greenlane Hospital, Dr Stephen Williams' cytological smear tests of sputum had been successful in detecting unsuspected lung cancer. He knew too that in New York Dr Papanicoloau, inventor of the 'Pap smear' was, to quote Williams, 'booming away about his cause'. Carey went to Williams and begged him to do a trial run of 1000 cervical smears of women entering the hospital for other reasons. 'Long before we had done 1000,' recalled Williams, 'at about the two hundredth specimen, we had turned up an extraordinary smear of carcinoma in situ with cancer cells spread right across the smear. There had not been the slightest sign of cancer of her cervix. Carey was dumbfounded. We did a biopsy tissue test of the cervix and discovered loads of CIS in the tissue.'

With this success, the programme did not stop at 1000 It continued and a full-time cytologist was appointed to set up a laboratory and train staff. He was sent to the United States to study cytology at first hand, and on his return the Papanicoloau grading sequence was introduced. By1957 the 'Pap smear' was offered to all doctors in New Zealand.

Carey was evangelistic in his belief in the benefits of screening. According to Williams: 'Carey raced round like a bee in a fit. He wrote an article for New Zealand Woman's Weekly saying that cervical screening could save five lives for every 1000 smears, a wildly exaggerated claim as it turned out.'

By 1964 the hospital cytology service was seeing 20,000 specimens a year and a new cytology block was put out to tender. But by this time Herb Green had stepped in and said he was having doubts.

Herb Green was born in Balclutha in 1916 and South Otago High School claims him as one of its most honoured old boys. At Otago University he obtained a BA and BSc, and finally qualified from Otago Medical School in 1945. For several years he represented the university at rugby.

Green joined the cervical cancer clinical team at National Women's Hospital in 1956. From then until his retirement in 1982 Green saw almost every woman who came to the hospital with invasive cancer, and many of those with the earlier or 'precursor' stages. Green told us it was experience with his own cases which led him to focus on cervical cancer and question established medical wisdom. A young poet had her university degree conferred in a ceremony in Ward 9 a week before she died of cancer of the cervix. 'When you see a young attractive woman who'd make a wonderful friend for a man a little younger than myself, when you see her die, it's not nice. That's why I've been so vocal. This is bad for mankind. I realised at the time she wasn't the only one under thirty, there were four.'

He took a scientific interest in the disease, and began writing papers about it. He travelled the world, attending conferences and taking up scholarships to further his study. The first paper on CIS appeared in 1962 and numerous papers appeared through the sixties and early seventies. It is very clear from the papers that Green was using as study subjects his patients at National Women's Hospital, for he was not secretive about what he was doing. However, although the medical world increasingly looked askance at his work, no criticism leaked out into public view.

By the early sixties, the idea of the progression of cervical cancer through precursor stages was generally accepted in the medical community. A later overseas witness to the cancer inquiry said: 'I don't recall a single one of my colleagues who from that time [1960] to the present did not regard carcinoma in situ as a precursor and who did not treat patients with this lesion.' The classic study by Petersen, reported in 1955, had provided compelling evidence. It showed that 33 per cent of 127 untreated women with abnormal cervical cells developed invasive cancer.

But although there was agreement that CIS progressed, how often and how quickly were the subject of debate. It was strongly argued through the sixties that population-based screening programmes would reduce the incidence of cancer and the large programme underway in British Columbia, Canada, was being watched with interest.

It was a patient, Shirley B, who gave Green the clue which formed the basis of his theory. He calls her 'the little girl who put me on the right track'. Green performed a biopsy, disclosing CIS, but Shirley 'refused a hysterectomy'. 'Two years later she had a baby. She's still alive, living somewhere around Whenuapai.'

In the fifties and early sixties the standard treatment for carcinoma in situ was hysterectomy, a major operation involving removal of the uterus or womb which therefore rendered a woman sterile. In the early sixties some hospitals began to use cone biopsy for young women. Cone biopsy is described as 'a nasty operation'. Haemorrhage needing emergency treatment is not an uncommon problem and the woman's chances of conceiving afterwards can be significantly reduced.

Green was concerned at any reduction in a woman's fertility. He was an opponent of abortion, and could see virtually no grounds for sterilisation, medical or social. Once sterilised, he said, a woman had 'thrown away a unique possession'. It was her 'heritage' to keep her uterus, Green told one woman.

Green became a man with a mission. He wanted to avoid sterilising surgery and to save women's fertility. To do so he had to prove what at first he had suspected, but then he came to believe: that CIS was a harmless disease which hardly, if ever, progressed to invasive cancer.

His belief led him to treat some patients less extensively. He stopped performing routine hysterectomies for women with CIS, instead performing cone biopsies. By 1962 he was able to report that no patients treated this way had progressed to invasion.

Having made a breakthrough and produced what looked like support for his hypothesis, Green pushed the limits of his theory further. Many other authors had published papers in the fifties and early sixties estimating that CIS progressed to invasion in a variable proportion of cases. Green proposed 'to show that the lesion is probably benign in the great majority of cases' and to question the view that screening programmes could eliminate invasive cancer.

Around 1960, Green began to treat small groups of women with CIS in unorthodox ways. At least five women, and possibly as many as sixteen, had hysterectomies without prior biopsies to diagnose what was happening to the cervix. Green had only looked at the cervix with the naked eye and taken a Pap smear. While a Pap smear can reveal an abnormality, a biopsy is necessary to accurately say what stage the disease is at. Earlier, Green had always argued that biopsies were vital for proper diagnosis. In these cases he omitted this step. One result was to provide him with intact cancer lesions to study in the laboratory. In Green's words: 'serial section studies of lesions thus undisturbed by biopsy are being made . . .'

There were two dangers in performing a hysterectomy without biopsy. Firstly, a healthy uterus might be needlessly removed. Secondly, and possibly fatally, if the woman had invasive cancer, it would be difficult to treat with radiation therapy after a hysterectomy. In this treatment, radioactive rods are inserted through the cervix to sterilise the uterus and adjacent tissue. With the womb removed, the rods could not be put in place.

Because in these cases Green had not done biopsies before surgery, the only way to check the seriousness of the disease was to immediately examine the excised wombs in the pathology laboratory. Before this could be done, Green took some of the wombs with him to New York when he left in 1962 on a Lederle Scholarship.

Some months later, National Women's Hospital received a cable from New York. Green was reporting that one of wombs contained invasive cancer.

The delay is significant. The original surgery would have spilled malignant cells into the blood stream. Speedy treatment usually follows a diagnosis of invasion, in this case the woman did not finally have radium treatment until six months after the hysterectomy. Despite the further surgical removal of part of this woman's vagina, she was past help and died two years later. She was thirty-four years old.

During the later cancer inquiry, this case became informally known, somewhat flippantly, as 'the flying uterus' case. It was discussed by a number of doctors and hospital administrators and was universally condemned.

In April 1963, Green embarked on another study. Between this date and June 1965 the rate of cone biopsy at National Women's Hospital doubled, peaking at 140 operations annually before falling to below forty in 1966. Wrote Green: '200 [cone biopsies] were performed by the author as a special study', one of the aims being to correlate what the prior smear said with the later histology [tissue diagnosis] report. Some of the post operative histology reports for these women showed they only had cervicitis, a mild inflammation of the cervix in no way requiring such drastic treatment.

In another group of patients the abnormal tissue was not completely removed, or Green 'temporised', that is, delayed further treatment despite positive smears continuing after cone biopsies. This, he acknowledged in writing, 'may appear foolhardy to those who believe in a high chance of progression of the disease to invasion'. Although it is clear from Green's writing that he knew his was a minority view, and that he was providing unorthodox treatment, he did not think the patient should be the one deciding: 'If we are uncertain about the natural history of the disease which cytology has revealed in her, how can we possibly expect her to make what is really our decision?'

Green justified his unusual handling of cases by insisting he always rigorously excluded the possibility of invasive cancer before deciding on limited treatment. He also produced his results in vindication of his methods. For much of the sixties he insisted that not a single case had progressed to invasion.

By the mid-sixties Green was so convinced CIS was harmless, he even argued against the necessity for cone biopsy, a proposition he decided to test on women. In correspondence with a GP Green expressed his thinking at this time: 'I must admit that the longer I go on, the bolder I get and we have now reached the stage where we are not even doing a cone biopsy in the younger age groups . . .' Later he wrote to the same GP that CIS and invasive cancer were 'totally different diseases'. Cases of invasive cancer appeared among women who had abnormal cervical cytology, he wrote, 'in much the same way as will happen in any large group of women followed because they had a broken leg'.

In June 1966 Green formalised his experiment and gained permission from the Hospital Medical Committee to withhold conventional treatment of cone biopsy or hysterectomy in women under thirty-five years old. It is quite clear from Green's proposal to the committee which we saw during the inquiry that he was already doing this with some patients. He already had, he said, eight patients who had not had cone or ring (a shallow cone) biopsy: 'All of these continue to have positive g smears at intervals.' There were also other women who had had a cone or ring biopsy which had not cured them: 'Many have had persistently positive smears including in one woman up to eight years following the original cone.'

The possibility of invasive cancer was excluded by punch biopsy performed by a colposcopist looking at the enlarged cervix. In a 1970 paper, Green explicitly stated his aim: 'to follow adequately diagnosed but untreated lesions indefinitely . . . to settle the question as to what happens to carcinoma in hi situ', and the criteria for selection. The women were to have no evidence of invasive cancer, such as intermenstrual or postcoital bleeding or bleeding from the cervical-canal when it was probed. The lesion had to be large enough not to be completely excised by the diagnostic punch biopsy'. Despite continuing abnormal positive smears, the women were to get no further treatment.

Other women who had abnormal smears after initial treatment such as cone biopsy or hysterectomy were also to receive no further treatment. According to a paper by Green, by the end of 1966 some seventy-three women with evidence of persistent, uncured CIS were being followed. Twelve of these had diagnosed micro- invasion of the cervix. Sixteen had only had a punch biopsy.

Differences of opinion over the best treatment for CIS were one thing. There was always unanimity in the medical community about the object of treatment - to return to a negative or normal smear. A positive smear was a sign that treatment had not been successful; there was still disease present and there had been no 'cure'. In such cases, further treatment was called for' until the smear became normal.

This was never the intention with the National Women's Hospital experiment. Some women with evidence of disease were to be left. They would be followed -that is, brought back for regular smears and possibly more biopsies - but there was no intention to cure them.

But by watching these women, Green hoped to observe the natural history of the disease and prove his thesis, that untreated CIS rarely if at all led to invasion.

Many of the overseas researchers would have liked to study the natural history of CIS, but it was considered unethical to conduct such a 'prospective' or forward-looking study. If you believed in the progression of the disease, as they did, you could not risk women developing the potentially fatal form by not treating its precursors.

In one of his papers, Green even quotes the famous New Zealand gynaecological specialist Professor Stallworthy, Professor of Obstetrics and Gynaecology at Oxford University, as having said in 1966: 'To regard carcinoma in situ as not being a killing disease is comparable to giving a stiletto to a paranoic.' In another paper Green quotes the opinions about progression held by others at the time; they varied from 33 per cent to 100 per cent.

Green was out on a limb. His opinion was not shared by the medical community and he knew it. A 1970 paper in an Australasian medical journal was headed 'An Atypical View'. This was insisted upon by the journal's editor. Green said that by 'commonly accepted standards, many in-situ lesions have been almost disdainfully under-treated', and that 'It is clear that the author's experience with cervical carcinoma in situ is at considerable variance with that of the great majority of others.' He called others' belief in progression a 'dogma' which had become 'almost unchallengeable' and so he set about challenging it.

Discussion of the ethics of the study never appeared in his papers. It never seems to have occurred to Green to worry about medical ethics because he believed so absolutely that women with CIS would not develop invasive cancer. Green wrote: 'If the physician does not worry too much about the disease, then neither will the patient.' A letter written by Green to a GP in 1965 gives an idea of the sort of message Green was conveying to his patients. The patient had only had a punch biopsy. 'Mrs G,' wrote Green, 'seemed prepared to accept my reassurance on the subject that she is in no danger of developing true cancer . . .'

In answer to a question from us about whether women were told there were differences of opinion about the methods of treatment, Green replied: 'I suppose not.' In answer to the same question, Professor Bonham said: 'I wouldn't know, you would have to ask each individual doctor who treated patients.'

Although the HMC, which passed the study, acted as an ethical committee, there was no hospital plan to seek the agreement of the women to their unorthodox management. Consequently, patients like Clare Matheson did not know they were being studied, nor that they were being treated in an unorthodox way.

For twenty years Bill McIndoe opposed what he saw happening at the hospital. He objected to the study from the moment it was mooted, predicting what would become the tragic reality within a few years. As the only hospital colposcopist, McIndoe was in the uncomfortable situation of being an integral part of a project he had serious doubts about from the start. As colposcopist for all the women involved, he knew he was observing the same lesions over and over again, lesions which in any other hospital would have been smartly removed.

McIndoe had originally trained as an electrical engineer after his teachers at Auckland Grammar said he was more suited to t work with his hands. After the war he enrolled at Otago Medical School, fulfulling a dream from his Bible camp days to be a doctor. He wanted to be a medical missionary, but by the time he qualified he and his wife had three daughters and it wasn't possible. Thus he came to medicine late, and at National Women's Hospital, where he took up a position in 1961, he never quite fitted in to the powerful inner clique which held sway at the hospital, the central figures of which were the professors, Bonham and Green. Doctors who worked hospital during this time actually talk of 'a Herb Green camp' and say that 'McIndoe dropped off quite early'.

In one of the hospital's operating theatres, McIndoe covered a disused colposcope, originally ordered by Professor Carey but never put into use. Colposcopy was a new technique pioneered in the fifties in Germany. McIndoe's training in optics at the university, his wartime work with range-finders and a lifelong interest in photography, naturally led him to be interested in the technique. In 1963 he went to Sydney to train in colposcopy under Malcolm Coppelson and the following year the Medical Research Council gave him a grant to further his work. He was also jointly in charge of the cytology department with Dr Stephen Williams.

Williams describes McIndoe as 'professionally very careful, exact and accurate. He was a worrier to some extent and at times he was in an enormous dilemma about all this business.'

According to Peter Mullins: 'He was worried sick about it. He could see people dying under their hands and that the standard wisdom being taught was wrong.'

McIndoe's obituary describes him as a 'kindly self-effacing man of complete integrity [who] could not abide sham or hypocrisy in any form. He was a good man whose high principles showed up strongly in his whole life . . .' A deeply religious man, he was, like Green, a Presbyterian and worshipped at the same Epsom church as Green's wife. He was a church elder, clerk of the kirk session for several years and leader of the senior Bible class for almost two decades. A Bible class member remembers McIndoe as someone who 'in large groups was not comfortable putting himself forward. He did not enjoy fighting and battling at all, even though he would do battle when he believed it was right to do so.'

It was this unlikely person who challenged the formidable figure of Associate Professor Herbert Green.

The first skirmish between Green and McIndoe was so low-key and muted that twenty years later, at the cancer inquiry, virtually nobody could remember it. The occasion was the meeting of the senior staff of National Women's Hospital on 20 June 1966. When McIndoe received the agenda for the meeting, he realised that Green was putting up a proposal for a trial about which he had extreme reservations and about which he, as the colposcopist involved, had not been consulted. McIndoe went to see the superintendent, Algar Warren, to discuss what he should do and to show a memorandum he proposed to bring up at the meeting

At the meeting, Green said his aim was 'to attempt to prove that carcinoma in situ is not a pre-malignant disease'. The time had come,Green argued, 'to diagnose and treat by lesser procedures than hitherto a selected group of patients with positive A3 - A5 smears'. All women 'under the age of 35 with positive smears in which there is no clinical or colposcopic evidence of invasive cancer' should be 'passed to the care of Professor Green whose conscience is clear and who could therefore accept complete responsibility for whatever happens . . .'

Punch biopsy specimens would be taken and 'if at any stage concern was felt for the safety of a patient a cone biopsy would be performed'.

The proposal makes it quite clear that women with positive smears were to be followed without any attempt being made to eradicate the abnormal tissue, but is otherwise woolly on critical matters, such as whose 'concern' would prompt surgery.

The minutes of the meeting record 'lengthy discussion' and that Green 'answered many questions'. There is no explicit mention on McIndoe's memo which he later recorded he tabled at the meeting. At the cancer inquiry, only Algar Warren, who chaired the meeting, could remember it.

In the memo, McIndoe argued against Green's proposal. The aim of treatment, he said, should be 'to aim to remove tissue responsible for the positive smear . . .' He called Green's proposal 'inadequate' from the point of view of diagnosis and of treatment.

The dangers McIndoe saw were twofold. Firstly, that by limiting diagnostic biopsies, a smouldering invasive cancer could be missed. Secondly, that by delaying treatment or not completely removing abnormalities, risks were being taken with women's lives.

At the later cancer inquiry, some of the doctors present that day argued that they bowed to the superior knowledge and experience of Green, the professor and expert. They were apparently unmoved, even unnoticing, of McIndoe's objections.

Mary Whaley is not surprised her father was unsuccessful. She describes the dynamics of the conflict this way: 'On one side were the people who made all the noise and did all the shouting, and on the other side was this shy little man in the corner saying he thought Green was wrong.'

Later on the day of 20 June 1966, Green's proposal went to the HMC, the elite group which controlled practices in the hospital. McIndoe was not a member. It was chaired by Professor Bonham and all the members had been present at the earlier meeting. At this meeting, Green's proposal was effectively simply rubber-stamped.

Although McIndoe was distressed about his failure to influence his colleagues, he thought for a time he might have been worrying unnecessarily. His relationship with Green was still relatively amicable; they were even co-authoring a paper on vaginal CIS. But by 1969, several cases of invasion had occurred in women who were part of Green's trial.

In the same year McIndoe was questioned about Green when he attended the First World Congress of Colposcopy in Rio de Janiero. The conference was attended by Leopold Koss and other pundits in cytology. They asked how Green achieved his results, results which were at complete variance with those of others working in the field. McIndoe had been asked this question everywhere he went: 'I was bound to state . . . that he is not presenting the evidence fairly! And, in fact, is biased in his reporting of what is actually happening in this hospital.' Some of McIndoe's remarks got back to Green, who tackled him about it.

There is no record of that conversation, but the next day McIndoe wrote Green a memo querying Green's practice of 'regrading patients originally diagnosed and classified as carcinoma in situ'. Green argued they had been invasive from the beginning. In this way they could not be counted as cases of progression and could be excluded from his published research. 'I have tried,' wrote McIndoe judiciously, 'to avoid the use of the word "progression" because it introduces a concept not helpful to these discussions.'

To Green he suggested that Malcolm Coppelson could come from Australia to see the work. 'We could all benefit,' he said. Green simply didn't respond to this memo at all. At the cancer inquiry he would say he never received it. McIndoe later wondered if he shouldn't have 'made my points with more vigour'.

Restrained the memo might have been, but-according to Mary Whaley it was 'the end': 'The battle lines were drawn.' In the memo, McIndoe listed seven cases where women originally diagnosed as CIS were later shown to have invasion. He began keeping lists of women who were 'causing concern'. Within two years, he had nineteen names. Some, he said, had been followed closely by himself and McLean 'for almost an unbearable length of time'.

He began keeping notes, meticulously recording events as they happened. According to Mary Whaley: 'He thought it would blow up. He thought there might be accusations levelled at him and that heads would roll. He wanted to keep his own name clear.' Although McIndoe did not live to see the final 'blow', at the later cancer inquiry many of us had reason to be grateful for his thoroughness.

McIndoe began raising with Green his treatment of individual patients, pressing him to bring them in for more extensive surgery. These approaches were seldom successful. Green, McIndoe and McLean would gather for histology sessions in McLean's room and increasingly these developed into rows.

It was the job of one of McIndoe's staff members to prepare a dossier of particular women. For some reason, a number of slides were kept in little cardboard boxes under the floorboards, and these would be retrieved and the woman's cytology history assembled. Along with McLean's histology, the whole case would be presented to Green and then,' she said, 'a confrontation would take place. McIndoe would come back in a lather about it and say he didn't think anything would happen.'

'On three occasions,' wrote McIndoe, 'I was invited to his room and subjected to intense interrogation. The first on a Monday morning, which lasted approximately three hours, at one stage was interrupted by Professor Liggins coming into the room . . . "Are you two still at it?" he said.'

In another 'particularly unpleasant episode,' McIndoe said, 'I was goaded into finally declaring him "irresponsible in his attitudes and management of patients".'

The arguments were about individual women, although there was after a time a depressing sameness about them. One that stuck in McIndoe's mind was the case of Mrs J. She had a histology report from a ring biopsy in 1968 confirming carcinoma in situ, but McLean gave a warning that the excision was incomplete and that the tumour could have invasive cancer nearby. Green did not act on this, and despite persistent A5 smears did nothing for three years, at which time he did another small biopsy. This showed invasive cancer, but still Green did not believe it. Three months later he did a cone biopsy. This time the report came back establishing invasive cancer, but Green said 'there is no evidence on clinical grounds that this woman has or ever has had an overt carcinoma of the cervix'. In other words, because he couldn't see it, he didn't believe it.

She had now had two histology reports from Dr McLean warning of invasive cancer, but Green disbelieved the diagnosis and planned a simple hysterectomy rather than combining it with the usual treatment with radium.

The exchange about this case which McIndoe remembered so vividly took place in the foyer of the cytology department. 'With a somewhat unusual and "challenging" look on his face,' he recorded, 'Green said he was going to do a simple hysterectomy on Mrs J . . . I made no forceful comment, simply shrugging my shoulders and saying, "Well, it's your decision", or words to that effect.' Even though the histology from the excised uterus showed 'extensive' cancer, Green did no follow-up radio-therapy. 'I do not think the final diagnosis of this patient is by any means certain,' he wrote. 'Histologically it may have been malignant, but clinically it has never been other than benign.' When another doctor wrote on her file that she had invasion, Green wrote: 'We shall see'. Unfortunately, time proved Green wrong, as it did in so many other cases. By the year of the cancer inquiry, this now elderly woman was still alive, but she had probable invasive cancer of the vaginal vault and invasive cancer of the vulva.

By March 1971 there had been a death. A full year had elapsed between the date this woman's punch biopsy had shown CIS and the date she was brought back for a more extensive biopsy which showed invasive cancer. In the intervening year, Green had noted the cervix had 'a very florid erosion and in fact would be regarded as suspicious by many people who don't look ` at a cervix very often . . .' When he performed the next biopsy, iP7 the cervix crumbled under the forceps. 'I think this is almost an invasive cancer of the cervix we have been sitting on,' noted Green. Despite a Wertheims hysterectomy, the woman died a little over a year later when the cancer metastacised to her bones. She was thirty-three years old.

This was the case Bonham had described as a 'colposcopic miss'.

When he tackled Green, McIndoe felt he rarely made an impact: 'I have not been able by commonly accepted methods to cause him to modify his long-held views and only occasionally to influence him to alter his planned management of particular patients.'

The response, said McIndoe, was that Green 'not only will not, but does not, listen to any comment which does not suit him. I have endeavoured by all means possible to make my feelings plain . . .'

'I can only assume . . . that I have never been sufficiently -belligerent and offensive for my comments and concern to be noticed.'

'It is very difficult,' McIndoe wrote, 'to maintain a high standard of reporting of the cytology work in such a climate. What can I say to a technician who misses an obviously malignant slide at the screening stage who replies, "What difference does it make anyway? They are not going to take any notice of the result."?'

In 1971 the National Women's Hospital Tumour Panel was established with Bill McIndoe as convenor. (Jock McLean had suggested it to the HMC while Green was overseas as a way of opening up discussion of particular cases of gynaecological cancer.) By this stage, McLean had joined McIndoe in his campaign. McIndoe's unhappy role in colposcoping the same untreated women over and over again was matched by McLean's experience in his laboratory. He was aware he was ' viewing the same tumours in the same women year after year. When he reported that the specimen showed CIS, or microinvasion, sometimes even occult invasive cancer, he was aware these alarming reports were not always acted upon. In other hospitals, such reports would have resulted in prompt treatment.

McLean says that junior resident doctors came along to the Tumour Panel meetings to be entertained by hearing Green, McIndoe and McLean slugging it out. As far as McLean was concerned this was 'progress': 'It meant controversial cases could be brought out into the open and discussed. It protected the patients and ensured they could be treated in an orthodox manner.' But the Tumour Panel was not a complete answer. Cases could only be discussed one at a time. It was, said McLean, 'like Lord Shaftesbury. We could not get everyone out of the mines at once.'

In 1973 McIndoe decided to take more formal action. The final spur to do so was the visit of the eminent Professor Per Kolstad of the Norwegian Radium Hospital in Oslo, who had years of experience with cancer. Six patients were examined by Kolstad, who, said McIndoe, 'made it clear that he would treat all the patients seen by total excision of the observed lesion'. Kolstad said more. He called Green's methods, bluntly, 'experimenting with patients'. McIndoe's task became urgent.

Eight years had elapsed since the institution of the study and his attempts to stop it had met scant success. He had taken it up verbally with the superintendent, Algar Warren, on several occasions. Three or four times, Warren did broach the subject with Green, but got nowhere. Green, said Warren, was difficult in every possible way.

Warren knew of the arguments between Green and McIndoe: 'Green came along to me once or twice in high dudgeon.' McIndoe, he said, despite being 'not a forceful character' nevertheless 'got under Dr Green's skin'.

Warren usually told McIndoe he and Green must resolve their differences themselves for he regarded himself as only 'the administrator'. McLean too had taken differences of opinion to Warren and received the same answer, although Warren saw McLean as more able to 'look after himself'.

When McIndoe brought him his formal memo of October 1973, Warren recognised that 'things had got to a head'. McIndoe was asking for 'a reappraisal of policy' for CIS patients. There were women, he said, who because of 'lim,ited biopsies' were in danger of having hidden invasive cancer go untreated. He outlined seven cases to illustrate his point.

McLean's accompanying memorandum to the superintendent outlined fourteen cases where limited treatment, he said, was 'not soundly based'. All the women were eventually shown to have invasive cancer. McLean contended they had probably had it from the beginning but because of inadequate biopsies it had been missed. 'In my opinion,' he concluded, 'these patients have had what can be termed varying degrees of delayed and inadequate treatment for a disease (invasive carcinoma considered to warrant urgent hospital admission for adequate diagnosis and definitive treatment.'

With McIndoe and McLean, Warren attempted to take the complaints outside the hospital to the controlling authority. They approached the medical superintendent-in-chief of the Auckland Hospital Board, Fred Moody, who discussed it with the board's chief executive. Despite the fact that the complaint came from the medical superintendent and two such senior doctors, Moody declined to step in. 'He threw it back to the HMC,' remembers Warren, 'because that was charged with the clinical responsibility.'

Moody explained: 'I had delegated my responsibility to thoroughly responsible and trustworthy superintendents and it was their job to do it, not mine.'

Although the memoranda from the two doctors ran to several pages and included details of cases of serious mismanagement, involving two deaths, Moody said there was 'insufficient evidence' and an 'absence of documentation'. He did not, however, think it his role to call for more documentation: 'This was a clinical issue and until the clinicians had had a very good look at it and made up their minds what they wanted, I could not interfere.' Instead, he set out in what Warren admiringly calls 'a magnificent memorandum' the structure within the hospital and the lines of responsibility as he saw it. If the HMC should not be able to resolve the issue, said Moody, the subject should be referred to the full hospital staff meeting. If the matter was still unresolved, but only then, it could be referred back to him. This was an unexpected setback for McIndoe and McLean. They had taken the unprecedented step of taking a complaint about a colleague to the hospital board, and it had not resulted in prompt action. McIndoe's wife, Noelene, recalls: 'I will never forget how depressed he was when he came back from the meeting with the superintendent-in-chief. He was totally despondent.'

In his reply to McLean and McIndoe, Green insisted that other clinicians were also following his mode of management and that other staff were implicated in the policy of limited treatment. He said that by studying the natural history of the disease 'a calculated risk that invasive cancer could be overlooked' was also present, but he denied that delayed treatment affected survival rates. For early invasive cancers, he said, the ten-year survival rate was 96 per cent. Patients whose cancers were 'overlooked or treatment delayed' were not 'at a disadvantage'.

In this war of memos, McLean and McIndoe penned further salvos, with McIndoe now able to list 100 women whose cases were causing him concern. McLean took Green up on his contention that the women involved hadn't really suffered: 'A survival rate of 96 per cent for Stage la carcinoma of the cervix is very good at a statistical level - but not for the four or so women who died from the disease . . . there is every possibility that with earlier diagnosis and treatment, these four or so patients could be alive.' He concluded with a clear statement of the issues at stake for the women: 'Women patients are admitted to a public hospital . . . with the explicit understanding that they will be provided with at least adequate, and preferably optimal treatment for their complaint. Clinical studies and trials to establish optimal management are at times necessary. However, when in the course of a trial it becomes apparent that patients are risk, there must be a reappraisal of the trial.'

The HMC set up an investigation committee of three senior doctors. Bernard Kyle, a senior HMC member, declined to be appointed, arguing that 'it is most inappropriate that this matter should be adjudicated upon by members of this hospital staff because of personalities and the fact that local staff have had a long time in which to prejudge the issue.' He suggested the appointment of a gynaecological pathologist and oncologist (cancer specialist) from another hospital.

Kyle's recommendation was not heeded and the other appointees did not share his qualms. They were all hospital doctors. The group was to be chaired by Mr A. Macfarlane, and the members were Bruce Faris and Richard Seddon, now professor of O&G at Otago University.

The terms of reference of the group were specific and narrow. The committee was only to look at the case notes of certain cases and decide whether 'they were managed in accordance with' the 1966 policy. They were not to carry out the reappraisal of the trial wanted by McLean. The committee finally began work in mid-1975, some twenty months after the visit to Moody.

The committee was initially given twenty-nine cases to consider, but for reasons which no one can now remember, considered only fourteen. The 'flying uterus' was then excluded because she was, according to Seddon, 'outside the period of investigation'.

In examining the remaining thirteen cases, the committee only asked whether the agreed 1966 policy had been followed in each individual case. It found that only three were within the under-thirty-five age group imposed in 1966. The age limit was set because women over the age of thirty-five would be unlikely to want further children; thus fertility was not such an issue. The significance of this age restriction also relates to the changing nature of the cervix as women age. In young women, the cells which can become malignant are located on the visible portion of the cervix. As a woman gets older, these vulnerable cells retreat up the cervical canal, out of the line of vision. In older women, malignant changes could be taking place without being detected by a smear, or colposcopic examination.

Without permission, Green had been ignoring the age limitation. Some of the women the committee considered were fifty- six, sixty-four and seventy-eight years old.

But despite the persistent violation of this rule, the committee determined that 'in twelve of the thirteen pertinent cases the agreed policy was followed'. Only the case of Mrs F (refereed t o earlier in this chapter), which had so upset McIndoe, was judged by the committee to have been not played according to the rules.

The committee noted delays of up to three and a half years after McLean had suggested invasive cancer in his histology reports and treatment had been instituted. It also noted that in all the cases invasion had been subsequently diagnosed. However, on that point the committee would not comment, for 'it is not within the terms of reference to comment on the outcome of the clinical trial'. Thus the committee effectively washed its hands of responsibility for twelve cases of invasive cancer appearing during the course of a trial. It was not moved to sound a warning bell.

Far from it. The committee decided that 'all staff members involved . . . have acted with personal and professional integrity'. Green's name is not mentioned; the only individual who gets an implied rap over the knuckles for faulty medicine is a 'junior medical staff member' for, it was alleged, 'inaccurate unconfirmed pathology reporting'.

The committee chastised the doctors for allowing 'differences of professional opinion' to develop into 'conflicts'. It does no t suggest that the trial be terminated, indeed 'the effective continuation of this trial,' it said, 'depends upon the staff members concerned subjugating personality differences in the interests of scientific enquiry'. It merely suggested a 'clarification' of the 1966 policy to iron out some of the rough spots. Not to safeguard the patients, but because they were 'a potential . . . source of disquiet between staff members'.

McLean had told us that Warren regarded the Macfarlane Report as a 'whitewash'. In the witness box Warren was reluctant to use this term but agreed the committee had 'dealt with the issue by ignoring the problems and upholding the status quo'.

The report of the Macfarlane Committee did not result in action, or even 'clarification' of the 1966 policy. There were only calls for yet more reports from McIndoe, McLean and Green. It was now almost three years since McIndoe had sought formal action. With a terrible sense of deja vu he set about writing again.

There is no point in detailing the substance of the reports of McIndoe and McLean. They were merely repeating, again, what they had been saying for the past five years. The only difference was that the casualty lists and the death tolls were rising, reflecting the fact that during all this time no action had been taken over the women. Green still retained his patients, and as the details provided by McIndoe showed, over the period of deliberation of the Macfarlane Committee, had not significantly altered his management techniques. Other clinicians stopped referring new CIS patients to Green, but the existing patients remained his responsibility, and his alone.

Dr McIndoe's submission lists in detail, over fifteen pages, sixty eases which had been 'under review' for some time before invasion was diagnosed. Most had had continuing positive cytology for some years. He instanced nine deaths. In 1973, when he first formally took action, there had been only one.

McLean's report talks of forty-three women who in the course of follow-up were found to have invasive cancer and twelve deaths. Six years before the cut-off date for the 1984 paper, McLean could instance more cases of invasion and more deaths than were finally included in that work. This highlights a fact that would become more apparent as the cervical cancer inquiry undolfed: the 1984 paper only painted part of the final tragic picture.

McLean drew attention to the ethical issues, pointing to , PrincipIe Seven of the Helsinki Declaration, the international code governing medical research, which he noted was endorsed by the New Zealand Medical Research Council. Principle Seven relates to an investigator's responsibility to discontinue a reearch project when it became apparent there are hazards to subjects.

Clearly, as far as he was concerned, the experiment was still going on.

In Green's report of July 1977, he admits that 'colposcopically directed punch biopsies cannot adequately exclude invasion', a conclusion he has reached by trying his proposition on women - over twelve years. He suggests a return to more cone biopsies, even for Grade 2R smears, an enormous shift from his previous position that cone biopsies were unnecessary, even for CIS.

However, he says his recommendation should be seen in the light of his 'personal feeling that the whole subject of cancer is far too emotive to both medical and lay people, and that this emotional outlook has vastly overrated the sociological importance of this disease . . . It is far too high up the priority list in medical circles.'

Green now had a new theory for the cause of cervical cancer. There might be, he said, 'two forms of cervical cancer . . . one .` with a long latent period which may be preceded by CIS . . . [and] one with a very short latent period, probably not preceded by CIS, which may develop explosively with no chance of being detected "early" by cytology'.

This 'two-cancer' theory provided Green with an explanation for the invasive cancer cases that occurred in the women under his care, and also provides him with a means of avoiding the more obvious conclusion, that his theory that CIS was benign had resulted in invasive cancer developing in a number of women.

He had expressed the same view to us during the interview when he said: 'You appear to cure the slow-growing ones, and you don't the fast-growing ones. Those were the lesions that couldn't be cured.'

Green makes it clear that he was still studying cases of CIS; he had nearly completed the coding of all 1044 of the hospital's cases. Peter Mullins, who had been a statistician at National Women's Hospital in 1976, worked with Green at analysing data on the hospital computer. He says he 'learned early that' you produced data and didn't discuss it. There was no point in talking to Green, you simply couldn't discuss scientific validity.'

Although warned of this by his predecessor, Mullins at first attempted to argue with Green about requests for investigations which were scientifically unsound. He discovered that this was a waste of time: 'He was the authority. He was the professor, so you ended up doing what he wanted even if you thought it was useless.'

To McIndoe it was critical that the outcome of Green's study be published. For over a decade, Green had bombarded the medical press with the results of his research, results which appeared to show that CIS was relatively harmless and the orthodox treatment methods unnecessarily drastic. It is one of the tenets of science that even disappointing and negative results should be published. When Green's results became overwhelmingly disastrous, he stopped publishing. Without an update and re-analysis of the cases, the medical world would be left with the wrong impression. Green's long-term results needed to come out, if for no other reason than that they proved him resoundingly wrong.

In mid-1978, the HMC finalised a policy on the treatment of CIS at National Women's Hospital. As far as the hospital authorities were concerned, this was the end of it. The process Moody had outlined in his 'magnificent memo' had been followed. It had just taken an awfully long time. But peace had been restored. Professor Green was thought to be doing more cone biopsies and Dr McIndoe was for the first time in many years quiescent.

But McIndoe was not really happy with the result. There had been no censure of Professor Green. He had not been instructed to stop experimenting on patients. By 1977 there was a formal Ethical Committee, chaired by Professor Bonham, but it had not formally terminated the 1966 study. Green had not been told to bring his patients into agreed clinical management methods. There had been no review of the cases of his existing patients.

Patients whom McIndoe knew from his records in the cytology department still had evidence of continuing disease had not been recalled. There was no hospital plan to bring them in and give them 'the treatment they needed. Peace had been restored at a price: a price which the women would pay, and no-one else

Noelene McIndoe confirmed that her husband continued to be obsessed with the problem. She remembers a car trip from Auckland to Wellington in the early eighties. McIndoe talked about it non-stop for nine hours. He talked about it compulsively at home. She took up croquet because 'he was away every Sunday going through the records. He never stopped reading and writing. It totally took over his life.'

There was not a great deal of support in the hospital for the work which would become the 1984 paper. In a letter to the medical superintendent in late 1982, Professor Bonham questioned what was happening. 'I have heard a rumour,' he wrote, 'that Dr Mclndoe and possibly another specialist, have been reviewing cases of carcinoma in situ that have been managed in the hospital. I have no recollection of approval being given

for review of in situ cases belonging to other consultants and I wonder if they have been reviewing cases by courtesy of Miss Owen [the secretary of the consultation clinic] without the approval of the clinicians concerned.

'This may only be a rumour but I think it may be worthwhile your having a look at it in the first instance because any publication emanating from this hospital must be acceptable to the staff of the hospital before it is submitted to publication, as I'm sure you will agree.'

From the other side of the world, Dick Mattingly, the editor of Obstetrics and Gynaecology, kept encouraging Mclndoe. 'While I fully understand,' he wrote, 'the medical politics involved in collecting the data, I cannot emphasise enough the importance that I, and others, place upon this singular study.' Mattingly knew, as did McIndoe, that the study was unique. No one had conducted a comparable study of this size.

'I'm sure you know,' continued Mattingly, 'how important this information is to the field of gynaecological oncology and how imperative it is to have this data published as a matter of scientific record. Without publication of your article, this "experiment of nature" will remain buried in the files of National Women's Hospital.

In this he was wrong. Publication of the 1984 paper failed to attract very much attention. It was only when Phillida and I brought the sad events at National Women's Hospital to the attention of the lay public that 'the experiment of nature' came out of its grave.