« AnteriorContinuar »
present. Especially is this explanation applicable where none of the above local complications appear to be prominent. In such instances the febrile miction often seems strictly analogous to the feverish condition producible by starvation, exposure, or over-exertion, in persona otherwise healthy. And in consonance with this view, I think it will generally be found that its presence is a rough test of the progress of the malady—at any rate of its effect upon the constitution; and is therefore not only one element of a very unfavourable prognosis, but an indication for as much nourishment and support as the local circumstances of the malady will allow.
The state of tlie bowds in gastric cancer is another point on which I have no exact numerical detnils to offer. In most instances, however, either constipation or diarrhoea is specified as having been present during a considerable period of the malady. And in many the two states alternate with each other; constipation in the earlier months of the history being succeeded by diarrhoea towards the last few weeks of life.
Constipation, the more frequent of the two, appears to be due to the circumstances of the malady rather than to the malady itself, and in this respect is comparable with the constipation which so frequently accompanies gastric ulcer. Obstruction, vomiting, and pain, which prevent alike the distension and the movements of the intestinal canal, concur to form a ready explanation of the frequency and intensity of this symptom, which sometimes persists up to the last hour of life.
Diarrhoea seems mainly attributable to the direct irritation caused by the cancerous, purulent, or haemorrhagic effusion which the gastric lesion introduces into the intestines. The noxious influences of these fluids are often aided by their decomposition or incipient putrefaction in their passage through the bowels. Ulceration of the Cfecum, and of the neighbouring ileum and colon, is comparatively so rare, as to afford Lut a very exceptional explanation. The date of the occurrence of diarrhoea is generally consonant with the above view of its production; it is subsequent to ulceration, in other words, is comparatively late in the history of the malady. It thus coincides with a period at which the patient's strength is generally so far exhausted a» to be very susceptible of any further depressing influence. Henceit often indicates imminent danger, and is not unfrequeutly the proximate cause of death.
The symptoms which more specifically betray the approach of death, offer little that is peculiar to this disease. Anasarca is by no meansuncommon, and is usually accompanied by ascites. A disproportionate effusion into the belly points rather to some local cause of the dropsy, such as pressure of the cancerous tumour on the vena porta:, and is hence less significant for prognosis. Jaundice, as before mentioned, appears sometimes to be a precursor of death, as is also a form of pulmonary effusion which I believe to be a mere passive congestion. Hiccough, again, is usually due to local causes acting on the diaphragm. The nervous phenomena that usher in the close of the malady am delirium, followed by coma.
The duration of the aggregate malady formed by the succession of all these symptoms may be calculated, from 198 cases, as averaging about 12^ months. This average seems to differ little in the two sexes. Ont of 142 cases, 107 male and 35 female, the average in the male was 12^, in the female 111- months. This slight difference is rendered even less important by the smaller number of females referred to, especially in conjunction with the fact, that the deviations from the average are far more striking and less trustworthy, in tho case of the longer durations than of the shorter. Indeed, as regards the longest or most chronic instances, it is curious to notice how few gradations exist between protraction to the close of the second year, and tho one or two rare examples of a duration of six or seven years. Hence these extremely protracted cases are not improbably examples of dyspepsia ending in cancer, an opinion which may also be directly deduced from their symptoms.
Practically, then, we may deduce that the maximum duration of gastric cancer amounts to a period of about thirty-six months from the first appearance of the symptoms, and that very few cases survive the twenty-fourth month. The minimum is about one month. As might be inferred from the histological details before alluded to,* many of the most protracted cases arc shown by necropsy to have a structure often designated "colloid" and "encephaloid," though originally scirrhus. The converse exception is much rarer, a large majority of the cancers rapidly fatal really belonging to these two intense forms of the malady.
But in noticing these facts, we must be careful not to overrate their significance. The commencement of the symptoms often has little connexion with the commencement of the disease. Pathologically, of course, it is not only absolutely impossible to fix the date of the deposit in any given instance, but probable that its material origin may be followed by all those varieties in the date of subsequent growth which we can verify in external cancers. And even that aggregate of symptoms which, for diagnostic purposes, we may make the representative of the disease, must often give a very distorted view of the real duration of the case. Slight anorexia, cachexia, or emaciation would often elude the notice of tho patient; or, perhaps, would rarely (in hospital practice) cause him to seek professional advice, such as might sometimes detect and distinguish the latent cause of these earlier symptoms. Hence, in most of the cases on which the above estimate is founded, the symptoms from which the commencement of the malady has been dated are those of active gastric disturbance in the shape of pain and vomiting.
This brings us to a point on which I am anxious to be explicit, in order not only to compensate the probable bias of the above estimate, but also to explain the deceptive resemblance between cancer and ulcer of the stomach—a resemblance which it is often of vital importance for the physician to unmask. In many oases of gastric cancer, the symptoms only date from the commencement of ulceration in the » British and Foreign Medico-Chirurgical Review, No. xxxvii. p. 22c.
cancerous deposit:—from a time, that is, when the incurable and malignant disease offers a close analogy, in many of its circumstances, to the curable ulcer; and an analogy which necessarily brings about the closest similarity in many of the local symptoms. As a rule, it seems that, whether we explain their access as due to this cause, or view them as independent of it, the more specific and frequent varieties of vomiting and pain, as well as the occurrence of hemorrhage, begin at a period about midway in the history of the best marked cases. Hence we may conjecture (1st) that the cancerous deposit itself generally precedes the symptoms by a period of at least some months; and (2nd) that the above estimate of the total duration of the symptoms will probably receive some extension from future clinical researches.
It may be interesting to combine the estimates we have come to respecting the frequency of the above symptoms separately, in order to determine their bearing on the diagnosis of the malady, or on the probabilities of any given case. Of the two estimates we have generally adduced for each symptom, we had better keep to the larger, both as a more direct and positive one, and as seeming to balance most exactly the two kinds of error—excess and deficiency—to which all such calculations are liable :—
Anorexia. Vomiting. Pain. Hemorrhage. Tumour. Cachexia.
Present 91 ... 193 ... 189 ... SI ... 128 ... 3 times, In . . 115 ... 221 ... 205 ... 202 ... 159 ... 4 cases.
A pplying to these figures the simpler laws of probabilities, brings out the following statements.
The probabilities or " odds" are eleven to two against all six symptoms being simultaneously present in any given case. In other words, one or other of the six is absent in more than five out of six examples of the malady.
Supposing the least frequent—haemorrhage—to be absent, the odds are still as many as eight to five against the concurrence of the remaining five.
It is only the absence of a second symptom—for example, cachexia —that equalizes the chances, so as to render exactly even the probabilities for and against the concurrence of the remaining four—pain, vomiting, anorexia, and tumour. In other words, the simultaneous presence of four of the above symptoms is about the average amount of evidence hitherto detected in cases of gastric cancer. The numerical frequency of anorexia and cachexia (70 to 75) is so nearly alike, that it scarcely matters which we include in this four; save that, on the whole, the presence or absence of cachexia is less mistakeable than that of anorexia. Hence supposing, in any given instance, we find pain, vomiting, tumour, and cachexia are all present, we may feel that the diagnosis, even if uncertain, is as little so as, in the majority of cases, it is likely to be.
The chances of concurrence would of course increase as the number of symptoms decreased. Thus it is 23 to 12 that pain, vomiting, and tumour are all present; 52| to 12 that pain is associated with vomiting.
A more practical hint is derived from an inquiry into the probabilities of a concurrent failure or absence of symptoms. Thus the odds are 2 to 1 against the simultaneous absence of haemorrhage and cachexia; 2| to 1 against the failure of haemorrhage and anorexia; 40 to 1 against the absence of tumour and vomiting; 60 to 1 against the absence of vomiting and pain; 65 to 1 against the absence of tumour and pain; 225 to 1 against the absence of the more obvious and important symptoms of pain, vomiting, haemorrhage, and tumour; and lastly, about 2560 to 1 against the absence of all the six symptoms tabulated above.
In alluding to such calculations, however, we must remember how large a variation of probabilities would often be brought about by a slight alteration in the original estimates. Supposing, for example, that we adopted the lower of the two estimates mentioned for most of the above symptoms, on the assumption that they were absent in every case in which they were not mentioned, the difference, we have seen, would not amount to much for each symptom separately. But its results would of course diverge far more widely; so that, for example, the odds would be about 55 to 1 (instead of 11 to 21) against all six symptoms concurring in any single case. The uncertainty which attaches to all calculations founded on limited numbers of cases may seem still more obvious. But in even as few as 200, this element of contingent error becomes comparatively small. Indeed the actual collation of these 200 cases singularly confirms the above calculations.
A similar, though less exact, check appears to be afforded by observation with respect to the general results of the preceding inquiry—which suggests, I think, rather a more secure diagnosis than many excellent authorities have hitherto thought practicable in this disease. This suggestion has certainly been confirmed by my own observations. In other words, I have rarely been unable to diagnose a cancer of the stomach, and have generally found that where its presence seemed long doubtful, there it was, in reality, absent.
But in whatever degree the facts above collected may aid diagnosis, by establishing conclusions such as no scantier series of observations could have justified, still it is right to add, that grounds for a definite opinion in any single case will sometimes be wanting. As already pointed out, in the phenomena of gastric cancer, time is everything. Hence the physician who examines a chronic case, or (what is practically the same thing) sees the malady only in its early stage, must often remain in complete uncertainty respecting its nature, until the progress of the disease, increasing (in something like a geometrical progression) the number and distinctness of its symptoms, enables him to come to a decision. Indeed, it is hardly too much to say, that this peculiar mode of accretion of the symptoms may sometimes form a specific aid to our recognition of the disease.
My limits forbid any allusion to the diagnosis of the various complications of gastric cancer. As pathological incidents or events of the malady, they have already been mentioned elsewhere. Their symptoms arc, on the whole, less distinet than those of the ordinary diseases which affect the organs they respectively occupy. Indeed, the late period of the gastric malady during which they usually occur, brings them into a coincidence with a degree of prostration such as obscures these symptoms, alike in respect to their own prominence, and to their reaction on the organism generally.