The belief that effective control of a genetic disorder of PKU— has been achieved was electrifying. The bandwagon quickly took on a heavy load-several voluntary health agencies, several Federal agencies, and medical researchers, among others. "PKU" became a household word. Interest in screening for other metabolic diseases was heightened. The National Foundation for Neuromuscular Diseases called for a "genetic alert" to discover children with such rare defects as Wilson's disease, agammaglobulinemia, glucose-6-phosphate dehydrogenase (G-6-PD) deficiency, and hereditary angioneurotic

edema.

People soon began discussing a new problem: What would happen when salvaged PKU patients grew up and procreated? Would they then add to the pool of defective genes in the total population? This, however, was a deferrable problem to be faced later. Meanwhile, in relatively short order, 37 States, as of November 1966, passed PKU screening laws. The American Medical Association also endorsed screening programs, as had other groups, but it did not favor legislative requirements for them.

5

Thus, the consensus was that an excellent medical model was available for the management of PKU and that screening and treatment should proceed accordingly. Only here and there were voices raised in protest against the legislative approach. Even fewer scientific challenges were expressed as to the validity of the medical model. I use the word "expressed" deliberately, because as I explored the PKU program I was to learn that many who had doubts had not expressed them. I also learned that the source of consensus was traceable to a rather small group of people who were in substantial agreement among themselves. The same people appeared time after time on professional programs throughout the country. They reinforced each other in their convictions, sincerely and earnestly held.

I do not personally challenge the professional consensus, for I am not professionally qualified to do so. I have, however, explored the literature and have consulted with biomedical authorities in order to assure myself that I was not reading and inferring out of context. In brief, the professional challenge to the consensus has been growing ever louder. If the challenge is sustained, we're going to find ourselves with 37 or more State laws which were enacted prematurely and with some children who have been permanently harmed without proof of offsetting benefit.

The evidence of a faulty medical model I will now present, covers diagnostic validity and reliability, control and effects of treatment, and readiness of the PKU program to be put into a service status for the public.

The diagnostic finding of a high phenylalanine content in the blood does not in itself establish the infant as phenylketonuric. Neither does the finding of a low serum phenylalanine level necessarily predict the absence of PKU. Guthrie and associates reported that about 85 percent or more of presumed positives turned out to be false positives." Persumably, the false positives are eliminated through use of confirmatory tests, but the literature offers no assurance that existing tests, used generally, will exclude all false positives. To the contrary, high phenylalanine levels may be included by nonrelevant conditions,

Ed. note. References 1-46 are set forth in Exhibit 1, pp. 61–63.

including the presence of infection. Even heterozygotes, otherwise normal, may show high phenylalanine levels.

The blood phenylalanine test is given in the hospital before discharge of the newborn; usually when 4 days old. The recommended procedure is that all infants be tested again at age 4 weeks, so that false positives and false negatives might then be rechecked. What happens to those babies who were put on low phenylalanine diets very early? How does a later test reveal them to be false positives or false negatives if already on diet? What happens to non-PKU children falsely diagnosed so to be? It has been suggested that when put on low phenylalanine diets, their own body proteins break down through catabolic action to release blood phenylalanine intake whereas more would actually be needed?

While Dr. Guthrie has reported, on the one hand, that 85 percent of his initial positive screenings are false upon recheck, other investigators have found "an unacceptably high percentage of false-negative results." Dr. Charles W. Blumenfeld, chief of the department of pathology, Sutter Community Hospitals and Sacramento Clinical Laboratory, California, reporting for himself and colleagues, claimed that the Guthrie test has a potential of 53 percent false negatives. So, on the one hand the test pulls in too many babies and, on the other, it is claimed to bypass too many.

One of the problems with which we are here confronted is that the most meager data have been used in projecting the incidence of disease, the validity and reliability of diagnostic screens and the clinical value of the dietary treatment. The efficacy of the dietary substance used to feed PKU babies, Lofenalac, was originally demonstrated through experience with only six cases as a basis for obtaining FDA approval in 1958.o

The trouble is in the difficulty of collecting enough new cases in any one clinical center to provide an adequate statistical base for study. Thus, the last time I checked on experiences in the District of Columbia, which has a screening program, only one confirmed case had been found in 2 years.

The likelihood of false positives increases with the rarity of a disease in the population.10 This explains why, with the Guthrie test, 85 out of each 100 children tentatively tagged as phenylketonuric are found on recheck, not to be phenylketonuric after all.

Consider the reactions of the parents who are told that their children are or might be retarded and will require additional tests over a period of time which will narrow the field down to only a fraction of those screened out. How does the imparting of this information change their lives? What traumata are thereby induced and at what cost? Even as to children who are eventually adjudged to be PKU-negative, are the lingering doubts in parents' minds ever dispelled?

At the May 1965 meeting of the Society for Pediatric Research in Philadelphia, Dr. Samuel P. Bessman brought up his famous false syllogism:

If one were to say that a woman is a person with two eyes and therefore that a person with two eyes is a woman, one would be laughed out of his august assembly. However, if one says that a mental defective who has phenylketonuria has a high blood level of phenylalanine *** [and] therefore a child with a high blood phenylalanine level *** is a phenylketonuric with mental retardation in the future is just as silly as the primary syllogism."

Ed. note. References 1-46 are set forth in Exhibit 1, pp. 61-63.

Thus far diagnosis has been discussed as an either/or matter: negative or positive.

One of the more recent uncertainties is whether "Phenylketonuria" is a single disease or a spectrum or collection of many different diseases each of which would require its variant of treatment. This is a matter for more profound conjecture than it has been given, particularly in view of our tendency to identify diseases and their cures through rather simplistic models.12

Consequently, children who test out to be phenylketonuric, in terms of blood chemistry, may grow up to be normal in all respects. Under present screening and treatment procedures, they would be put on a diet which might very well do them harm. On the other hand, some children who develop normally on a PKU diet might have developed even more normally without the diet.

A factor which impairs the ability of scientists to conduct research, even to know whether or not the low phenylalanine diet is beneficial, is that when an entire universe-all babies who show up as PKU positives-is put on the diet, no control groups are left for comparison. We do not know whether any children under dietary management who do not become retarded would have remained normal under ordinary diets. Untreated children with high phenylalanine levels have been found with intelligence in the normal and high normal range.13-22 Knox reported that about 10 percent of untreated cases are only mildly retarded and that less than one-half of 1 percent have been reported as PKU-normals.23 How any one can report faithfully as to percentages of normals is not easy to understand, in view of the near-universal treatment of babies diagnosed as phenylketonuric.

To sum up this discussion, thus far: (1) We have diagnostic tests which may or may not correctly identify babies who need preventive treatment; (2) we determine the existence of a disease state from a single symptom-the level of phenylalanine, and (3) we treat babies who have that one symptom without really knowing whether we have excluded those babies who would grow up to be normal without

treatment.

In short, the evidence for diagnostic validity is shaky, at the very least. Yet treatment programs have been instituted throughout the United States and have been exported to other lands despite the lack of diagnosic refinement.

When I have raised questions about the diagnostic situation, I have been told by PKU program enthusiasts that the proof of the pudding is in the eating thereof. They point to their own experiences in obtaining striking improvements in behavior and intelligence of children on PKU-preventive diets. MacCready of Massachusetts said, for example, that of 36 children in his State screened in the period July 1962, through February 1966, of which 29 were treated at Boston Children's Hospital: "All of the children, whose ages range from 1 month to three and one-half years, are in good health. The I.Q.'s done, admittedly approximate, range from 80 to 138 with a median close to 100." 24 With such experience, on face so convincing, how can anyone challenge either the diagnostic or treatment procedures? The answer is that many have done so and that serious doubts are being raised increasingly as to the value of the dietary treatment and its safety.

Ed. note. References 1-46 are set forth in Exhibit 1, pp. 61-63.

Earlier, I mentioned the widely held belief that diagnosis and treatment within the first weeks of life are important in order to prevent brain damage. Some recent work by Fuller, a psychologist of New York, suggests otherwise. She found that, taking all phenylketonurics (within her study) as a group, the earlier they were placed on the low phenylalanine diet, the less impaired they were. This would possibly support the views on early diagnosis and treatment, but let us face that one further after looking at some more of Fuller's findings.

Some PKU babies started on early control showed marked retardation. On the other hand, some children started on the diet after 3 years of age-long after brain damage should have occurred-showed great improvement to the point of little or no retardation. Furthermore, in a study of siblings on PKU diets, a significant proportion started at a later age showed better improvement than their sisters or brothers who were started at younger ages.2

25

The implications are profound. It is not improbable that measured improvement is in part, at least, the result of an intervention factorthe giving of individual attention to children who previously had been abandoned by society-rather than to biochemical effects of withholding an amino acid from the diet. The intervention factor, which may in this case be response to attention and training, is known among psychologists as the "Hawthorn effect."

Fuller's findings of intelligence improvement did not agree at all with the Massachusetts experience. Both as to her early-treatment group and her late-treatment group (after 3 years), she found a median intelligence of 69, much lower than the IQ of 100 reported in Massachusetts. Another study of Massachusetts phenylketonuries by Kennedy and associates showed a mean IQ for "severe" phenylketonuricsics, after treatment of 104, with a range of 80 to 136. Mild phenylketonurics had a median IQ after treatment of 109 with a range of 90 to 127. Interestingly, the severer cases tested out poorer and better at the rang eextremes as compared with the milder cases.26

Clearly, someone ought to find out just why such sharp differences were found in IQ measurement in the two States. Maybe different environments and measurement methods would be found. In particular, I would like to know why so many PKU-children who are found to be high-normal after treatment do so poorly in school, as I have been informed.

Another problem area is the absence of information as to how long a child should remain on diet and the effects of dietary termination.27 Solomons and associates of Iowa found favorable results in a study of seven cases, suggesting that termination of the diet after age 6 "is safe and seems to produce improvement in behavior, school performance, and in the emotional climate of the family." 28 On the other hand, Fuller found some tendency toward deterioration.29

As may be gathered, I am deliberately presenting a picture of utter confusion which should be quite acceptable during an investigational stage of a new program, but not during an advanced stage of a universally accepted therapy and mandatory by statute. Later, I will go into the question of whether the public has been correctly informed as to the true reasons for and nature of the PKU program.

Ed. note. References 1-46 are set forth in Exhibit 1, pp. 61-63.

Thus far, I have dealt with inconclusiveness of data as a basis for action. Now I will take up problems of treatment, especially as related to child safety. Very little attention has been given to deleterious effects of treatment, including the incidence of death attributable to induced dietary insufficiency. Other side effects and complications of the PKU diet, which consists solely of an unpalatable preparation from which phenylalanine has been removed, include skin lesions, refractory anemia, bone changes, vomiting, lethargy, appetite failure, poor weight gain, retarded growth, and miscellaneous symptoms of malnutrition.30

Recently, investigators found that the immune responses of PKU children to certain infectious diseases were lowered by dietary treatment.31 Wooley has argued that PKU patients saved from idiocy might become schizophrenic when taken off their diets later in life.32 The Food and Drug Administration informed me it had no knowledge of deaths. The Children's Bureau informed me it had no evidence of fatalities occurring as a result of treatment. Yet references to treatment-induced death are indeed to be found in the medical literature.

The Committee on the Handicapped Child of the American Academy of Pediatrics, while concluding that children with PKU can be helped through early detection and treatment, also reported: "This conclusion must not lead to unrealistic expectations or to over-enthusiastic application of treatment programs. Some parents are either unwilling or unable to maintain dietary treatment. Over-rigidity of dietary management has led to early death, presumably from insufficient protein intake or hypoglycemia. Over-hospitalization for rigid control has deprived children of the normal stimulation and affection of home and family thus preventing normal psychological maturation. Exaggerated predictions for normal development regardless of the age of discovery and irrespective of the strictness of the diet or of the hereditary endowment have led to frustration and discouragement on the part of both pediatricians and parents." 33

38

Careful perusal of the literature reveals both general and specific references to death.34-37 MacCready and Hussey reported two deaths of PKU children, but attributed one to pneumonia and the other to suffocation. Subsequently, in an exchange of correspondence published in the Medical Tribune, MacCready revised the report, in the light of subsequent information, and acknowledged that at least one of the deaths appeared to have been dietarily induced.39

The difficulty in attributing cause of death is that the child dies of one of a number of conditions brought about by withdrawal of essential nutrients. He may consume his own tissues. He may die of anemia. He becomes prone to infection. How does the physician attribute cause on the death certificate? Is he likely, in any event, to record a death as resulting from something done to the child? He can suspect, but does he really know?

A reported case of death, as an example, was attributed to megaloblastic anemia of nutritional origin. A phenylketonuric infant was started on a low-phenylalanine diet at age 2 weeks, suffered serious setbacks resulting in three hospital readmissions, and died at age 7 months.40 The reference to this death was published in a bibliography of the Children's Bureau.

Ed. note. References 1-46 are set forth in Exhibit 1, pp. 61-63.