INTENSIVE VITAMIN THERAPY IN MEASLES BY J. B. ELLISON, M.A.. M.D.. D.P.H., ASSISTANT MEDICAL OFFICER, L.C.C. FEVER HOSPITAL SERVICE

Recent research seems to have crystallized into the belief that vitamin A is concerned with raising the resistance of the animal body as a whole, and the epithelial cells in particular, to various bacterial invasions. The comparative scarcity, of human investigations on this subject may perhaps be attributed to the fact that it is not easy to point to any clearly defined group of symptoms associated with a deficiency of this factor in man in any way analogous to the syndromes of rickets and scurvy. There is, however, good evidence for the belief that concentrates rich in vitamin A are valuable as prophylactics against infections.

The work of Mellanby and his colleagues¹ on the prevention of puerperal sepsis by the use of such a concentrate appears convincing within the limitations imposed by the comparison of relatively small samples. There is, however, less evidence that a favourable influence can be exerted by such means on the course of a disease already in progress. The results achieved by Mellanby and Green² in the treatment (as opposed to prophylaxis) of puerperal infections await further confirmation. In Canada, Wright and his colleagues³ failed to protect a small series of babies against the "common cold" by supplying them with large amounts of vitamin A, while those contracting the disease did not appear to escape more lightly than the controls. Nevertheless, we are not entirely without indications to guide us in the search for a profitable field for vitamin therapy. In his classical experiments on the deficiency xerophthalmia of rats in 1922, Mori⁴ observed atrophy of the mucus-secreting cells in the salivary glands and in the mucous membrane of the larynx and trachea. If vitamin A was withheld for a further period, a local bacterial invasion of these tissues followed. Mori observed that most of his animals eventually died from broncho-pneumonia. These findings were confirmed by Cramer i and Kingsbury,⁵ and similar observations have been made by Steenbock,⁶Mellanby,⁷ and others.

The changes which occur in epithelial cells following deprivation of vitamin A have been studied by Wolbach and Howe,⁸ and later by Goldblatt and Benischek,⁹ who were able to demonstrate metaplastic degeneration in columnar, cuboidal, and transitional epithelia, leading to the appearance of a squamous keratinizing type of cell. Mellanby and Green¹⁰ have suggested that the desquamation of such cells may provide a favourable medium for bacterial growth. In any case it is clear that serious damage to the ciliated epithelium and mucus-secreting cells lining the upper respiratory tract must render the lung alveoli easily accessible to bacterial invasion. Therefore it is in diseases in which the patient's resistance to his own epithelial saprophytes is temporarily in abeyance that we might expect on a priori grounds to see benefit from the intensive exhibition of the factor which normally retains the defensive cells in good repair. But there is one considerable difficulty to be encountered in any human investigations on this subject. We cannot compare the condition of a man with that of his brother who has been totally deprived of his vitamin A factor. The point to be decided is whether an adult—or, a fortiori, a child—with a very liberal supply of vitamin A is better off in an emergency than one who has perhaps just enough for his normal occasions.

Now a disease which attacks epithelial defences and whose incidence is greatest in those members of the community who are most likely to be suffering from various grades of vitamin deficiency will probably prove the best medium for testing the therapeutic properties of vitamin concentrates. Measles appears to me to fulfill these criteria. In measles the brunt of the attack falls on epithelial structures. The respiratory tract rarely escapes unscathed. Indeed, the condition of the lungs, trachea, and bronchi in a severe attack bears some resemblance to that which Mori observed in his devitaminized rats. The mucous cells are injured, the ciliated epithelium is shed, and a definite metaplasia is prone to occur in the stroma of the lungs.

Taking advantage of these opportunities, the normally harmless saprophytes, frequently headed by the bacillus of Pfeiffer, invade the lymphatics of the trachea and bronchi, producing in the lungs, through the irritation of their toxins, a further degree of hyperplasia of the inter-alveolar reticulum, easily recognized as the much dreaded "acute interstitial pneumonia" of measles.¹¹ At the same time the general immunity of the skin is lowered—hence the boils, styes, impetigos, eczemas, and other minor tribulations—or it may be lost completely, leading to the various forms of cutaneous gangrene which are among the major horrors of the disease. The whole alimentary tract is liable to serious catarrh, especially in the Tropics. The Eustachian tube, middle ear, and mastoid cells are liable to dangerous invasion.

In this country the disease afflicts most heavily the children of the poorest classes, among whom the greatest mortality is observed during the first eighteen months after weaning. Since most of them are suffering from a lack of suitable fats in the diet, it is natural to suppose that they are insufficiently furnished with vitamins A and D. In the present investigation an attempt has been made to decide whether a liberal supply of these two factors at an early stage is able to exercise a favourable influence on the course of the disease. Observations were made on 600 cases of measles in children under 5 years of age, admitted to the Grove Hospital between October, 1931, and April, 1932. These were divided into two equal groups. One received an accurately measured quantity of a rich concentrate of vitamins A and D from the first day of admission. The other received the normal diet supplied in measles. Naturally vitamins were not withheld from them, but they received no liver oils during the febrile stage, though the usual dose of cod-liver oil was supplied during convalescence. Thus the treated cases received a relatively enormous dose of vitamins A and D for a. limited period, while the controls received the usual amounts administered to sick children.

Each of the treated cases received 20 minims daily of "adexolin," a cod-liver oil concentrate at present on the market, either in liquid form or in capsules containing an equivalent amount of the vitamins. According to the manufacturers statement (Glaxo Laboratories) the daily quantity administered amounted to 300 units (Carr and Price) of vitamin A, and 2,000 international units of D— that is to say approximately the amount contained in one ounce of high-grade cod-liver oil. This dose was continued irrespective of age for a minimum period of seven days in mild cases and for a maximum period of three weeks in those developing complications.

It is essential before attempting to draw conclusions from such a relatively small sample of the total measles population to show that the two groups were strictly comparable in all respects. In order to secure a perfectly random distribution of cases between the two groups all the cases, admitted to certain wards between October, 1931, and April, 1932, received the concentrate, while all cases admitted to the remaining measles wards served as controls. It was necessary to ensure absolute congruence of the two groups in point of time, since it would be clearly unsound to compare cases treated in February with controls selected in June. It was also necessary to obtain a similar age distribution in both groups, because the age of the patient is the most important single factor affecting the prognosis of measles in London.

The age distribution and corresponding death rates in the two groups are shown in the following table:

These figures require more careful analysis before we can attribute the lower death rates observed in the vitamin treated cases to the effect of the concentrate. The observed difference of 5 per cent. in the total death rates is approximately 2.5 times the "standard error," or 3.7 times the "probable error" of simple sampling. The most that can safely be said is that in two random samples of 300 cases such a distribution is not likely to occur purely by chance.

A comparison of these figures with those obtained in the previous epidemic appears to increase their significance.

Table II.—Acute Fever Hospitals, October, 1929, to July. 1930 (L.C.C. Special Report, July, 1931)

Age	Number of Admissions	Deaths	Rate %
0-1	488	78	15.9
1-2	1,453	201	13.8
2-3	1,110	70	6.3
3-4	1,041	34	3.2
4-5	886	19	2.1
Total	4,978	402	8.1

It will be seen that there is a fairly close correspondence between the death rates of the large group of just under five thousand cases and the controls in the present investigation. A somewhat higher mortality was to be expected in my series, because mild cases occurring towards the end of the epidemic in May, June, and July were excluded. This comparison shows that the expected death rate at these ages lies somewhere between 8 and 9 per cent.; hence the rate of 3.7 observed in the treated cases appears unusually low.

These facts are made clearer by a consideration of the following table, which shows the number of cases admitted to hospital either already suffering pneumonia or developing the complication after admission.

(note.—The figures in bracket indicate the number of cases in which pneumonia developed after admission to hospital.)

This table accounts for all the deaths except four— three of the controls and one of the treated cases. These deaths were attributed to acute gastro-enteritis. The pneumonia attack rate of 10.8 per cent. for the total of 600 cases is lower than that usually recorded in measles during the same period of time. This is attributable to the fact that only those cases are included in which no possible doubt could exist concerning the presence of definite pneumonic areas after the disappearance of the rash. A number of doubtful cases have been excluded from both groups.

The distribution shown in Table III is somewhat unfavourable to the controls, though it is noteworthy that the number of cases presenting signs of pneumonia on admission to hospital was greater by one in the treated group than in the controls. It is, howver, obvious that the numbers are too small to permit any very fine analysis.

An attempt was also made to determine whether any protection of the middle car could be obtained with the concentrate. Only cases free from ear discharge on

Age	Controls	Vitamin Cases
0-1	2	2
1-2	15	12
2-3	4	6
3-4	6	6
4-5	1	3
Totals	28 (1)	29 (1)

It will be seen that no benefit whatever was obtained from the concentrate in preventing otitis media; nor was there any perceptible difference in the length of time , required for the otorrhoea to cease in the two groups. A negative result has also to be recorded as regards the various skin complications. The figures are too small to be worth recording in extenso, but it was quite clear that the vitamin cases did not escape any more lightly than the controls.

The dose of the concentrate administered in this experiment was quite arbitrary. It may have been greater than necessary, but it is not possible to state to what extent the vitamins were absorbed and assimilated. Contrary to expectation, enteritis did not prove any drawback to the treatment ; actually this complication proved less troublesome among the treated cases than in the controls. In the only case in which proved fatal from acute gastroenteritis in the vitamin group the complication did not develop until a fortnight after the withdrawal of the concentrate.

The youngest cases were carefully watched for any signs of hypervitaminosis, but this was never detected. In a preliminary experiment I have observed definite signs of intolerance in a marasmic baby who had received 6,000 international units of vitamin D daily for ten days. The symptoms were diarrhoea, anorexia, and marked irritability. Rapid improvement occurred on withholding the vitamin. From a study of recorded cases it appears unusual for clinical evidence of hypervitaminosis to develop in such a short space of time, but it is possible that this child had been receiving vitamin D in some form before admission to hospital. There is no reason to fear that 2,000 units given daily to a baby for two or three weeks will produce any undesirable effects. Vitamin A seems to be perfectly innocuous even in huge doses.

There are no grounds for supposing that the vitamin D in the concentrate exerted any specific beneficial effect on the course of the pneumonias. The work of Goldblatt and Benischek⁹ has demonstrated that only vitamin A can claim to possess any protective value againstrespiratory infections in animals, while recent clinic observations of Friedlaender¹² have shown that the presence of rickets does not seem to affect unfavourably the prognosis in bronchopneumonia. Nevertheless it is possible that some adjuvant effect was obtained from the co-operation of the two factors, since the concentrate "adexolin" is stated to contain them in the natural proportions in which they occur in cod-liver oil.

Although the evidence submitted in this experiment scarcely amounts to a demonstration of the value of intensive vitamin therapy in measles, it is difficult to resist the conclusion that some protective effect was obtained in the treated cases which tended to limit the severity of the pulmonary complications. The use of a concentrate rich in vitamin A as a prophylactic against secondary infections in a population of young children known to have been exposed to measles might well repay further study. Obviously too much cannot be expected in the way of rescuing cases already desperately ill with bronchopneumonia, since it is almost inconceivable that a sufficiently rapid effect can be obtained by such means.

The total failure to reduce the otological and cutaneous complications is disappointing, but here again it is reasonable to suppose that better results might have been achieved had it been possible to start the treatment during the incubation period, though it is unlikely that the problem will be solved by these means alone. In presenting the results of this experiment I am fully conscious of the dangers of founding arguments on the comparison of relatively small samples.

In a statistical investigation of this nature it is probably wisest to set down an impartial record of what was conserved, leaving the reader to draw his own conclusions regarding the significance of the figures obtained.

1. Three hundred cases of measles received a concentrate of vitamins A and D during the acute stage of the disease: eleven deaths occurred in the series. In a control series of 300 cases having a similar age distribution twenty-six-deaths occurred.

2. Evidence is brought forward in support of the view that the pulmonary complications were less severe in the treated cases than in the controls.

3. No difference was detected between the two groups regarding the number of cases developing otological or cutaneous complications.

My acknowledgements are due to Dr. J. H. Whitaker, medical superintendent of the Grove Hospital, for permission to make use of the clinical material at the hospital, and also; to the director of the Glaxo Laboratories for a liberal supply of the concentrate "adexolin."

1. Green, H. N., Pinder, D., Davis, C., and Mellanby. E.: British Medical Journal, 1931, ii, 595.

3. Wright, H. P., Frosst, J. B., Puchel, F.. and Lawrence, M. R.:Canadian Med. Assoc. Journ., 1931, p. 412.

11. Ellison, J. B.: Arch. Dis. Child., 1931, vi, 37. 12 Friedlaender, A.: Ugeskrift for Laeger. 1931, p 1174

Controls				Treated cases
Age	Number	Deaths	Rate %	Number	Deaths	Rate %
0-1	32	4	12.5	31	1	3.2
1-2	90	16	17.8	101	8	8.0
2-3	60	4	6.7	55	1	1.8
3-4	61	2	3.3	61	1	1.6
4-5	57	0	0	52	0	0
Total	300	26	8.7	300	11	3.7

Controls			Vitamin cases
Age	Cases of Pneumonia	Deaths	Cases with Pneumonia	Deaths
0-1	4	4	4	1
1-2	20	13	13	7
2-3	8	4	8	1
3-4	2	2	6	1
4-5	0	0	1	0
Totals	34 (5)	23 (5)	32 (2)	10 (2)