Palatal paralysis may occur in these patients and is indicated by nasal voice and regurgitation of the food through the nose.

We will now give the statistics in the 36 cases having very serious heart complications as indicated. by gallop rhythm. 19 were males and 17 females. The ages were as follows:

Two-thirds of these patients died, one on the second day, the others on the fourth to eighth days. The effect of late treatment upon the mortality is shown by the fact that fourfifths of the fatal cases had no treatment until after the fourth day. Among the cases which recovered three-fourths were treated before the fourth day. The severity of the illness. is indicated by the fact that at entrance all but two patients had a pulse of over 120 and several a pulse of 140 or over.

The first important cardiac symptom that we consider is gallop rhythm; this was found in all the cases. It appeared in one case on the sixth day, in one on the 27th, in all the others from the 9th to the 18th day. When gallop rhythm appears murmurs which usually are present become indistinguishable.

Late vomiting is the next important symptom. This occurred in about two-thirds of the cases. It was present in

all but four of the fatal cases; of these four cases three died of paralysis of respiration and deglutition in the sixth to the eighth week having recovered from the gallop rhythm, leaving only one patient who died during the gallop stage and who had no vomiting. Thus we see that late vomiting was almost invariable in the fatal cases. Among the patients

who recovered, on the other hand, there was only one with gallop rhythm who had also late vomiting. In nearly every case the appearance of the gallop rhythm preceded the onset of vomiting from one to four days, occasionally longer.

The next symptom is epigastric pain and tenderness. Nine-tenths of the patients who had this symptom died; of the two who recovered one had only slight tenderness and no vomiting, the other had well marked tenderness and pain with vomiting. The time of appearance of this fatal symptom varied; it usually was closely associated in time with the appearance of vomiting, either a day or two before or a day or two after.

Palatal paralysis was noted in 19 of these severe cases, sometimes before, sometimes after the onset of heart complications; its earliest appearance being the 9th day of the illness and its latest the 17th day. It was found in ten of the fatal cases. There is difficulty in recognizing its presence in very sick patients, and it probably was present but not observed in others.

We have said enough to show that gallop rhythm, late vomiting, epigastric pain and tenderness were very frequent in severe and fatal cases, and very rare in those who recovered, which is sufficient to emphasize their importance as danger signals and their value in prognosis.

The illness of the patients who recovered was protracted, -one-third of the patients remained in the hospital six to eight weeks, one-third from eight to ten weeks, and onethird from ten to twelve weeks. During their stay in the hospital all but one had marked paralysis of legs, arms and eyes. In the fatal cases the duration of life was short, all but three died between the tenth and twentieth day of their illness, three lived from six to eight weeks; all the latter suffered from late paralyses.

The condition of these patients at discharge was interesting. Among those who recovered, two-thirds had an

irregular pulse, all but one had murmurs at the apex and base of the heart, all but one had a weak first sound, all had a rapid pulse rate, the slowest being 110. In two cases which recovered the heart was enlarged at discharge; in practically all of the fatal cases the heart was dilated.

The other signs of cardiac disturbance were not striking; sometimes there was slight oedema of the ankles and scanty urine; the respiration was rapid, averaging 35, and often irregular, especially at the latter part of the illness; there was no cyanosis. One case discharged at the end of six weeks had no paralysis; all the others had severe paralyses.

The great danger in delaying treatment with antitoxin is emphasized by the occurrence of so many serious and fatal heart complications, the likelihood of prolonged or permanent damage to the heart even in cases which recovered, and the abundant and severe post-diphtheritic paralyses, for it will be remembered that four-fifths of these fatal cases had no treatment with antitoxin until four or five days after the illness began. The cases which recovered had no treatment for at least three days, and they proved to be long, severe and dangerous.

Treatment.—It has been our experience that rest in bed and nursing are most important in the treatment of heart complications and that drugs play a small part.

Our figures show that the serious complications nearly always develop within three weeks from the onset of the illness, although in rare cases they may appear several days later.

It is usually safe to allow the mild cases to be out of bed at the end of two weeks. The presence of murmurs and a slight degree of irregularity are no contraindication if the first sound is strong and the heart is not dilated. Many of the cases severely ill, or ill several days without treatment, apparently do well for two weeks or more, the pulse may be a little irregular and rapid, but the condition of the

patient causes no anxiety. It is not, however, safe to allow these cases out of bed, even if the pulse is normal in rate and of fair strength. If four or five weeks pass, and no serious heart complications arise, the danger from this source is usually over, although death may occur from other late effects of the disease, such as paralyses of respiration and deglutition. It has been our experience that, after four or five weeks, a patient with well marked murmurs and irregular pulse, even if rapid, experiences no ill results from being allowed to sit up if the heart is carefully watched.

We will separately describe the treatment in ordinary cases, in those with persistent rapid pulse and in those with gallop rhythm. In cases with a murmur and little irregularity the treatment is directed to improving the general condition of the child. The patient is allowed to sit up after two or three weeks of illness, but active exercise is forbidden. Digitalis and alcohol rarely do any good; strychnia is the best drug, and is given in doses of onesixtieth of a grain every four hours for a child of five to ten years. In many cases larger doses seem to be useful, and produce no unfavorable effects.

Cases with persistently rapid pulse should be kept in bed at least four or five weeks and then allowed to sit up for short and progressively longer intervals, the heart being carefully watched on each occasion. Frequently, after four or five weeks rest in bed, the heart action is improved by this change. Strychnia seems to be the only drug of any value.

In cases with gallop rhythm absolute rest in bed is necessary, and a liquid diet is given. At the first appearance of vomiting, food should be given by rectum and cracked ice given by mouth for thirst. Strychnia in doses of onesixtieth to one-fortieth of a grain every four hours for children three years and over is given with the best results, although drugs influence this condition but little.

If there

is vomiting, the strychnia should be given subcutaneously. Cocain in doses of one-twelfth grain every four to six hours in ice cold water at times relieves the vomiting; alcoholic drugs, even champagne, are not well borne, and often increase the vomiting. Morphia in doses of one-sixteenth to one-eighth grain, depending on the age, is of value to quiet restlessness. Predigested beef products, if given in small amounts, are often retained when everything else is rejected. Hot applications to the epigastrium sometimes relieve the vomiting and pain. Digitalis is useless, and usually increases the vomiting.

Sudden deaths have occurred in our series of cases from asphyxia, but there have been no sudden deaths in the hospital from heart complications. It has occurred several times, however, that patients who have been taken out against advice about the end of the second week have died suddenly during the stage of gallop rhythm when allowed out of bed. For example, a child of three years after three weeks' stay in the hospital was taken home against advice just at the time in the illness when serious heart complications are common. A few days later the child died suddenly while being held up to the window to see some friends. This emphasizes the necessity for absolute rest in bed until all likelihood of the development of gallop rhythm has passed. The rest does not prevent the appearance of the complications but renders the outlook relatively favorable.

The cases of bradycardia require no special treatment beyond rest in bed and strychnia. Long duration of murmurs and irregularity of the heart indicate the necessity for following cases of diphtheria long after convalescence to study and prevent possible permanent changes in the heart.

The after treatment of all except the mild cases consists in watching the effect of mild exercise upon the heart for several months at least, and grading it to meet individual requirements.