Metabolic Studies in a Case of Adrenocortical Hyperfunction Associated with Carcinoma of the Lung
R. D. COHEN, I. P. ROSS,1 AND A. D. DAYAN
Medical Unit and the Departments of Medicine and Pathology, The London Hospital, London, E.1, England
ABSTRACT. Sodium and potassium bal- ances and indices of acid-base metabolism have been studied in a patient with carcinoma of the lung associated with adrenal cortical hyperfunction and hypokalemic alkalosis. Alkalemia persisted in spite of the retention of
1300 mEq potassium. After initial fluctuations the urine pH remained persistently between 5.2 and 6.0. The possibility that factors other than potassium deficiency were responsible for the paradoxical aciduria is considered. (J Clin Endocr 24: 401, 1964)
T HIS PAPER describes metabolic studies made in a patient suffering from carcinoma of the lung with adreno- cortical hyperfunction and hypokalemic alkalosis. A notable feature was the per- sistence of alkalosis during retention of 1300 mEq of potassium (administered as the chloride). In spite of the alkalosis the urinary pH remained between 5.2 and 6.0. Possible explanations for this finding are considered.
Methods
Conventional balance techniques were used. Intake of sodium and potassium was assessed by chemical analysis of duplicate diets. Occasional variations from the diet were estimated from the tables of McCance and Widdowson (1).
The following chemical methods were employed: plasma, urinary fecal, and dietary sodium and potassium, flame photometry (after wet-ashing diets and feces); arterial pH, standard bicarbonate and pCO2 (2); plasma bicarbonate, Technicon Autoan- alyzer; mixed venous pCO2 (3); arterial oxy- gen saturation, Brinkman hemoreflector; urine pH, Pye pH meter; urine ammonia (5), urinary 17-keto and 17-ketogenic ster- oids (6), urine aldosterone excretion (7),
Received October 17, 1963; accepted January 10, 1964.
1 Present address : Southend General Hospital, Southend, Essex, England.
urine amino acid nitrogen (8); plasma and urinary creatinine (9).
Case History
The patient was a 64-year-old man. He had been subject to constipation all his life and for many years had taken a nightly dose of aloes and podophyllin, but had not suf- fered from diarrhea. His sister was known to have renal tubular acidosis, with nephro- calcinosis and hypokalemia due to renal potassium loss.
He was first admitted in March, 1959, be- cause of 8 months of unexplained vomiting. Physical examination was negative, chest roentgenogram was normal, and plasma electrolytes were as follows: sodium 154, potassium 5.2, chloride 112, bicarbonate 32 mEq/1. A laparotomy was performed but no abnormality was found. He was then symp- tom-free till April, 1962, when he had a transient episode of severe weakness of the legs. In July, 1962, his ankles began to swell, and he was given chlorothiazide for a fort- night. Shortly after this he again noticed some weakness of the legs, which rapidly increased, and on August 12 watery diarrhea began. He was readmitted on August 14. Further recent symptoms included weight loss of 6.5 kg, nocturia X1-2 and cough. On examination he was ill, wasted and slightly drowsy (B.P. 140/70). He appeared dehy- drated and had an acute left parotitis. There was no edema. The urine contained a cloud of albumin, no sugar and no significant de- posit. Weight 47.7 kg (13th day after admis- sion).
| Initial Investigations | ||
| Plasma Na | 138 mEq/l | |
| K | 2.0 mEq/1 | |
| C1 | 72 mEq/1 | |
| HCO3 | 30 mEq/l | |
| Urea | 60 mg/100 ml | |
| Plasma urea | (17) | 69 mg/100 ml |
| (21) | 79 mg/100 ml | |
| (33) | 78 mg/100 ml | |
| (37) | 64 mg/100 ml | |
| 24 hr creatinine | ||
| clearance | (10) | 65 ml/min |
| (26) | 61 ml/min | |
| Plasma chloride | (7) | 87 mEq/1 |
| (11) | 88 mEq/l | |
| (14) | 106 mEq/ml | |
| (20) | 97 mEq/1 | |
| (44) | 108 mEq/1 | |
| Arterial O2% saturation | ||
| (7) | 85% | |
| (9) | 89 % | |
| (15) | 85% | |
| (18) | 88 % | |
| (36) | 91 % | |
| Urinary | ||
| 17-ketosteroids | (29) | 27.4 mg/24 hr |
| (30) | 26.4 mg/24 hr | |
| (36) | 40.6 mg/24 hr | |
| (37) | 34.0 mg/24 hr | |
| Urinary 17-ketogenic | ||
| steroids | (29) | 26.5 mg/24 hr |
| (30) | 24.2 mg/24 hr | |
| (36) | 53.0 mg/24 hr | |
| (37) | 28.3 mg/24 hr | |
| Urinary aldosterone | (17) | <1 ug/24 hr |
| Plasma albumin | (8) | 2.4 g/100 ml |
| (16) | 2.6 g/100 ml | |
| (36) | 2.9 g/100 ml | |
| Plasma globulin | (8) | 2.4 g/100 ml |
| (16) | 2.2 g/100 ml | |
| (36) | 2.0 g/100 ml |
The figures in parentheses refer to the day of the study on which the measurement was made.
Investigations and Progress. The biochemical features on admission are shown in Table 1. The initial diagnosis was dehydration and hypokalemia, probably due to diarrhea and chlorothiazide therapy, with suppurative parotitis and aspiration pneumonia. Chest roentgenogram showed only minimal sha- dowing in the right mid-zone. He was rehy- drated and received 260 mEq of potassium. On the 7th day he developed profound weak- ness, shallow respiration and twitching, and
lost consciousness. Arterial blood revealed standard HCO3, 51 mEq/1; pCO2, 57 mm Hg; pH, 7.61; oxygen saturation, 85%; and plasma potassium, 2.0 mEq/l. The electro- cardiogram showed changes compatible with potassium depletion (S-T depression, Q-T., 0.63 sec). The urine potassium concentration was 28 mEq/1, and, since his daily output of urine was nearly 3 liters, the urinary loss was thought to be the main cause of the deple- tion. The urine pH was 7.25. After 10 mEq of potassium chloride intravenously in 30 min he regained consciousness and was able to take potassium supplements orally. At this point the balance studies shown in Fig. 1 were begun. Potassium repletion was carried out with potassium chloride in the first 12 hr, effervescent potassium tablets (citric acid 1 part, potassium acid tartrate 3 parts, po- tassium bicarbonate 5 parts) for 3 days, and potassium chloride thereafter. The daily potassium intake is shown in Fig. 1 and varied between 80 and 320 mEq/day. Despite strongly positive potassium balance his weakness increased, with marked proxi- mal wasting suggestive of carcinomatous myopathy. Serial roentgenograms eventually revealed a mass at the right hilum, strongly suggesting carcinoma. Glycosuria became heavy and persistent by day 20, though without ketonuria. Repeated glucose toler- ance tests revealed an increasingly diabetic curve, and the diabetes was controlled by Lente Insulin, 40 units daily, starting on day 38. Daily blood pressure readings during the first fortnight of his admission were normal, but later rose to 200/110 to 180/130. By day 37 the electrocardiogramabnormalities attributed to hypokalemia had almost com- pletely resolved (Q-Tc =0.42 sec). The urinary excretion of 17-keto and 17-keto- genic steroids was found to be elevated (see below); at no time, however, did he present the clinical appearance of Cushing’s syn- drome. He died on the 49th day after admis- sion.
Biochemical Studies (Fig. 1). Potassium balance was positive, providing the potas- sium intake was greater than 80 mEq/day. The total potassium retained was about 1300 mEq by day 35. Fecal potassium measure- ments were not available after that date, but, from the data shown in Fig. 1, it seems likely that he was in approximate potassium balance, especially as some of the urinary creatinine measurements were somewhat
low in the last week, suggesting occasional incomplete urinary collections. The plasma potassium slowly rose; between days 34 and 45 it was within normal limits. In the last 2 days of his life, when potassium intake was poor and uncertain in amount, the level fell to 3.5 mEq/l. The abrupt rise of plasma potassium to 5.6 mEq/l on day 15 occurred soon after an increase of potassium intake to over 300 mEq/day and also with the admin- istration of Aldactone-A, 50 mg 3 times a day. This drug was given because at that stage primary aldosteronism was considered as a diagnosis and it was hoped by giving
this aldosterone antagonist to decrease the renal wastage of potassium. However, no fall in the urinary potassium occurred, and when Aldactone-A was given again from day 26 to day 35, while potassium intake was approxi- mately constant, there was no change in the plasma or urinary potassium. The rise of plasma potassium on day 15 was therefore probably due entirely to the increased intake.
The sodium balance was strongly positive for the first 10 days of the balance, coinciding with a weight gain of 2.5 kg, increasing edema, and a transient rise in jugular venous pressure; thereafter it became negative. The
50 mgm. t.d.s ALDACTONE A
50 mqm.t.d.s. ALDACTONE A
6 qm NHẠCI
5 qm NHẠCI
4 7 qm NHẠCI
8
7
URINE pH
0%
6
ph.Art. 7.6
pCO,mmHg
5
60
60
7.5
50
50
PH
NORMAL ARTERIAL RANGES
7.4
40
40
pCO2
I
7.3
30
Std. HCO, meq /l
30
Standard HCO,
PH pCO2
160
PLASMA Na+ meq/l
I St.HCO,
140
120
6.0
PLASMA K + meq/l
4.0
2.0
100
Na balance meq.
0
100
0
K balance meq.
100
200
20
URINE
300
0
PHOSPHATE meq/day
100
URINE NH+
0
meq/day
7
12
17
22
27
32
37
42
47 1
Days after Admission
DIED
7
80
RATE OF EXCRETION OF NH4 + DEq min
URINE
60
6
PH
40
5
20
NH4 CL
4
0
0
2
4
6
0
2
hours
8
4
hours
6
8
100
60
70
.
50
·
RATE OF EXCRETION OF TITRATABLE ACID
NH4+ RATE OF EXCRETION DEq
40
40
30-
uEg
20
min
20
min
0
0
2
4
6
8
4
5
6
7
hours
PH
plasma sodium exhibited large fluctuations, approximately coinciding with alterations in plasma potassium.
The arterial pH was 7.61 during the epi- sode of coma and dropped to 7.47 after 6 days of potassium repletion. It then rose to 7.51 and thereafter remained at this level (nor- mal arterial pH, 7.35-7.42). The standard bicarbonate, initially 51 mEq/1, fell after 5 days to 30 mEq/1, rose to 36 mEq/1, and then slowly fell to 29.5 mEq/l (normal standard bicarbonate, 21.3-24.8 mEq/1). The arterial pCO2 slowly fell from its initial high level of 57-63 mm Hg to within normal limits (34-45 mm Hg). The urine pH rose in the first few days of repletion from 7.2 to 8.0. When the arterial pH fell to the region of 7.5, the urine pH also fell abruptly. A single dose of 6 g of ammonium chloride was given on day 9 without change in the patient’s clinical con- dition. From day 14 onward most urine pH measurements lay between 5.2 and 6.0. On day 23 a standard acid load was given (4.9 g ammonium chloride, i.e., 0.1 g/kg body wt), and the urinary pH and rates of titratable acid and ammonia excretion were followed (10) (see Fig. 2). A minimum urine pH of 5.17 [within the normal limits reported by Wrong and Davies (10)] was reached after 4 hr; subsequently the pH was slightly above the upper limit of normal; the titratable acid excretion was normal initially, but after 4 hr fell slightly below normal. The rate of am- monia production was normal, though more
ammonia was produced for a given pH than normal. The titratable acid excretion on day 48 was 23 mEq/24 hr. During the last 10 days of the study the urine phosphate excre- tion was usually about 20 mmols/day, mostly as H2PO4-, since the urine pH lay between 5.2 and 6.0. The ammonia excretion averaged about 65 mEq/24 hr during this period.
The results of urinary steroid estimations and other serial measurements are shown in Table 1. The urine amino-acid nitrogen was raised (197 mg/g creatinine). Chromatog- raphy demonstrated an abnormal pattern of amino acids in the urine, threonine and alanine being moderately increased, and cystine and tyrosine considerably so.
Autopsy. Oat cell carcinoma in lingula; sep- arate epidermoid carcinoma in right middle lobe, with metastases in right hilar lymph nodes. Hypertrophic adrenals (each 12 g); compact cells resembling those of zona reticularis occupied almost the entire cortex, which showed occasional microadenoma formation; no distinct zona fasciculata; zona glomerulosa represented by occasional clumps of 5-10 small cells; small metastasis of oat cell carcinoma in right adrenal medulla. The anterior pituitary showed many Crooke cells and an excess of large chromophobes, with lightly granulated cytoplasm forming in one place an indistinct nodule. Other endocrine glands normal. A few small oat cell meta-
stases in the liver. Detailed macroscopic and histological examination of the brain, includ- ing hypothalamus and cerebellum, showed no abnormality. The kidneys (weights: R., 141 g, L., 150 g) also appeared normal, and there was no evidence of pyelonephritis or vacuolar degeneration. Nonspecific myo- pathic degeneration in sections of diaphragm, right deltoid and quadriceps.
Discussion
The association of carcinoma, fre- quently bronchogenic, with Cushing’s syndrome has been the subject of a num- ber of recent papers (e.g., 11-13). Our patient did not have the appearance of Cushing’s syndrome, and in this respect resembled 14 of approximately 55 pa- tients in the literature. The case history demonstrates the importance of repeated chest radiology and measurement of urinary 17-keto and ketogenic steroids in the diagnosis of renal potassium loss of uncertain cause. The principal feature of interest on investigation was the be- havior of the acid-base balance during potassium repletion.
We attribute the raised levels of arterial pCO2 at the beginning of the metabolic studies to hypoventilation compensatory to alkalosis and possibly also to the myopathy affecting the res- piratory muscles. Later in the study the pCO2 fell to within the normal range de- spite the persistently elevated arterial pH. The extent of compensatory hypo- ventilation may be limited by the degree of anoxia produced, and perhaps by a lowered intracellular pH in the respira- tory center (14). Our patient, in fact, had considerable arterial oxygen desat- uration even at the beginning of the study when chest roentgenographic changes were minimal.
The acid loading test is somewhat difficult to interpret because the normal range refers to subjects without alkalo- sis. It was performed while the patient
was still potassium-depleted. The classic disturbance of acid excretion in potas- sium depletion is failure to establish the normal maximum hydrogen ion gradient between blood and urine after an acid load (19, 20). It is possible that the dose of acidifying salt given in this case may have been partially neutralized by the alkalemia, and that a completely normal result might have been obtained if a larger dose had been administered. Sim- ilarly, the near-normal result of the acid load test in the presence of alkalosis makes it unlikely that the patient had renal tubular acidosis, as did his sister.
The blood urea levels were somewhat elevated and creatinine clearances were slightly reduced. Because the kidneys appeared histologically normal at au- topsy it seems probable that the lowered glomerular filtration rates were func- tional in origin and not due to pre-exist- ing renal disease. The raised blood ureas may be attributed partly to the rapidly progressive muscle wasting and partly to the depression of glomerular filtration.
The initial potassium depletion was severe, amounting to at least 1300 mEq. The fact that the patient was closely approaching, if not in, potassium balance during the last ten days of his life (see above) does not necessarily mean that he was fully potassium-repleted; but the consistently normal level of plasma potassium at this time, and the large amount of potassium retained, make it likely that he was in fact approaching repletion. In spite of this, his extracellu- lar alkalosis persisted. Many workers have demonstrated that the plasma HCO3- falls rapidly on starting potas- sium repletion (e.g., 15, 16). Cooke et al. (16) showed that this was related to a shift of hydrogen ions from the cells to the extracellular compartment. Aber et al. (14) and Atkins and Schwartz (17) presented evidence suggesting that any
chloride deficiency had to be made good before administration of potassium would abolish the extracellular alkalosis of potassium depletion. Our patient re- ceived potassium mainly in the form of the chloride, and the plasma chloride level rose to high normal values, so that, although chloride balance data are not available, it is unlikely that the persis- tent extracellular alkalosis can be at- tributed to chloride deficiency.
The excretion of acid urine in the presence of an extracellular alkalosis is well known to occur in potassium de- ficiency. It has been suggested that this paradoxical aciduria is due to the distal tubular cells sharing in the generally low intracellular pH caused by potassium depletion (18). They are thus prevented from “recognizing” the extracellular alkalosis, and hydrogen ion is secreted. In our case the administration of potas- sium in the first few days of the balance study may have allowed the tubules to recognize the alkalemia, for a strongly alkaline urine was at first excreted. As a result of this and the release of hydrogen ion from the intracellular compartment, the arterial pH fell from 7.61 to 7.51. Thereafter, the urine pH remained usu- ally between 5.2 and 6.0. It appears that the renal excretion of acid was reset to maintain the abnormally high arterial pH, which remained elevated despite the fact that large quantities of hydrogen ion must have been available since at least 80 mEq/day were lost in the urine in the last days of the study (calculated from the ammonia and phosphate excretions).
The high arterial pH and acid urine persisted unchanged during the retention of large amounts of potassium. Further, a review of the literature reveals that the pH of the urine in potassium deficiency is nearly always above 6.2; in our pa- tient, from day 14 onward the urine pH was usually lower (5.2-6.0). It therefore
seems possible that the paradoxical aciduria was not due entirely to potas- sium deficiency at any stage. A definite conclusion cannot be reached on this point because of the uncertainty as to how much potassium deficit, if any, re- mained at the end of the study.
If the disturbance of the acid-base regulation described in this paper is not purely due to potassium depletion, other explanations must be entertained. Its possible relation to adrenal hyperfunc- tion must be considered. Mills, Thomas and Williamson (21) and Bartter and Fourman (22) have shown in acute ex- periments that glucocorticoids such as hydrocortisone or prednisolone lower the total renal excretion of hydrogen ion, whereas aldosterone and desoxycortico- sterone increase it. Bartter and Fourman (22) also showed that acute injection of ACTH had an effect similar to that of the administration of glucocorticoids. It is not possible to deduce from these acute experiments what the long-term effect of these steroids would be. It is conceivable that the abnormality of acid-base metab- olism seen in our patient could be due to an unusual pattern of circulating ster- oids, and it is of interest to note that Cost (23) has recently demonstrated much higher levels of corticosterone in adrenal hyperfunction associated with nonadrenal neoplasms than in classic Cushing’s syndrome. Hyperaldosteron- ism is unlikely to play a part in our case, since renal potassium loss was not affected by Aldactone and the urinary aldosterone was very low. Furthermore, normal or low aldosterone excretion and normal secretion rates have been demon- strated in other cases of Cushing’s syn- drome associated with neoplastic disease (11, 13). In view of the tubular abnor- mality in the sister of this patient, it is possible that a familial abnormality existed which affected the response of
renal acid-base handling to excessive adrenocortical activity.
Acknowledgments
Our thanks are due to the late Lord Evans for permission to publish this case, to Professor Clifford Wilson, Professor I. Doniach, Dr. J. M. Ledingham and Dr. A. S. Mason for helpful dis- cussion, and to Mrs. E. Scott, Miss G. Low, and the laboratory, nursing and dietetic staff of the Metabolic Ward. We are grateful to Dr. R. Simpson for performing some of the arterial estimations, and to Miss E. Aitken for the aldosterone estimation.
References
1. McCance, R. A., and E. M. Widdowson, The Composition of Foods, H. M. Stationery Office, 1960.
2. Astrup, P., K. Jorgensen, O. S. Anderson, and K. Engel, Lancet i: 1035, 1960.
3. Campbell, E. J. M., and J. B. L. Howell, Brit. Med J i: 458, 1960.
4. Conway, E. J., Microdiffusion Analysis and Volumetric Error, ed. 2, Crosby Lockwood, London, 1947.
5. Taussky, H. H., and E. Shorr, J Biol Chem 202: 675, 1953.
6. Norymberski, J. K., R. D. Stubbs, and H. F. West, Lancet i: 1276, 1953.
7. Ayres, P. J., O. Garrod, S. A. Simpson, and J. F. Tait, Biochem J 65: 639, 1957.
8. Clarkson, T. W., and J. E. Kench, Biochem J 62: 764, 1956.
9. Bonsnes, R. W., and H. H. Taussky, J Biol Chem 158: 581, 1945.
10. Wrong, O., and H. E. F. Davies, Quart J Med 28: 259, 1959.
11. Allott, E. N., and M. O. Skelton, Lancet ii: 278, 1960.
12. Bagshawe, K. D., Lancet ii: 284, 1960.
13. Meador, C. K., G. W. Liddle, D. P. Island, W. E. Nicholson, C. P. Lucas, J. G. Nuck- ton, and J. A. Luetscher, J Clin Endocr 22: 693, 1962.
14. Aber, G. M., P. A. Samson, T. P. White- head, and B. N. Brooke, Lancet ii: 1028, 1962.
15. Black, D. A. K., and M. D. Milne, Clin Sci 11: 397, 1952.
16. Cooke, R. E., W. E. Segar, D. B. Cheek, F. E. Coville, and D. C. Darrow, J Clin Invest 31: 798, 1952.
17. Atkins, E. L., and W. B. Schwartz, J Clin Invest 41: 218, 1962.
18. Milne, M. D., R. C. Muehrcke, and B. E. Heard, Brit Med Bull 13: 15, 1957.
19. Clarke, E., B. M. Evans, I. McIntyre, and M. D. Milne, Clin Sci 14: 421, 1955.
20. Eales, L., and G. C. Linder, Quart J Med 25: 539, 1956.
21. Mills, J. N., S. Thomas, and K. S. William- son, J Physiol (London) 151: 312, 1960.
22. Bartter, F. C., and P. Fourman, Metabolism 11: 6, 1962.
23. Cost, W. S., Lancet i: 362, 1963.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName} ] on 14 November 2015. at 22:35 For personal use only. No other uses without permission. . All rights reserved.