Bartter’s Syndrome, 1962
Hyperplasia of the Juxtaglomerular Complex with Hyperaldosteronism and Hypokalemic Alkalosis, A New Syndrome
Frederic C. Bartter M.D.
HYPERPLASIA and hypertrophy of the juxtaglomerular apparatus, and primary aldosteronism with hypokalemic alkalosis were found associated with normal blood pressure in two patients  . This coexistence of a histologic lesion not previously described with a rare disease entity suggests an association between them; the appearance of these disorders in two patients who have never had hypertension appears to represent a new syndrome.
We here report studies designed to elucidate the pathologic physiology of this syndrome. The findings provide strong support for the growing body of evidence that adrenal cortical secretion may be influenced by the renin-angiotensin system.
Patient C. J. was a five year old Negro boy who had been admitted to Children’s Hospital, Washington, D. C., with complaints of tetany and dwarfism, at the age of four years and ten months. He weighed 8 kg. and appeared dehydrated. Carpopedal and quadriceps femoris spasms were present, and Chvostek’s sign was positive. The serum calcium was normal, but the tetany would respond temporarily to calcium gluconate therapy. Hypokalemia was found, and he was referred to the National Institutes of Health for further studies.
Full term normal delivery had followed an uneventful pregnancy. His birth weight was 3 kg., his birth length 42 cm.
At the age of four months he was hospitalized because of fever of one week’s duration, associated with vomiting, diarrhea, dehydration and generalized convulsions. A lumbar puncture, skull roentgenograms and gastrointestinal series revealed no abnormalities; slight albuminuria was present. A pneumoencephalogram showed slight dilation of the lateral ventricles. Since discharge he had been well except for retardation of growth and polydipsia which required him to drink 10 to 12 glasses of fluid daily. His family history was non-contributory.
* From the Clinical Endocrinology Branch, National Heart Institute, and the National Cancer Institute, National Institutes of Health, Bethesda, Maryland.
We are indebted to Dr. Joseph Lo Presti for referring this patient.
Physical examination revealed a short (36 inches), proportionately dwarfed Negro boy (span = 37 inches) who weighed 8 kg. and appeared moderately dehydrated. The blood pressure ranged between 92/52 and 120/80 mm. Hg on numerous determinations with an average of 100/60 mm. Hg. Chvostek’s sign was positive and carpopedal spasm readily appeared when he cried. Physical and neurologic examinations were otherwise non-contributory.
Numerous urine specimens showed a pH of 7 or above, specific gravity 1.010 or below, occasionally a trace of albumin, and no sugar.
Chemical analysis revealed the following serum values: Potassium 2.2 mEq. per L., carbon dioxide content 34 mEq. per L., chloride 75 mEq. per L., sodium 130 mEq. per L., and blood urea nitrogen 15 mg. per 100 ml. The result of a serologic test for syphilis was negative.
Repeated electroencephalograms all showed “diffuse dysrhythmia.” Several records also showed “bilaterally synchronous paroxysmal activity of the atypical centrencephalic type.” The electrocardiogram revealed a “T-vector of low magnitude, rotated posteriorly (juvenile pattern), S-T segment depression and prominent U waves in leads V2 through V4 .” Roentgenograms revealed a bone age of three years; and an intravenous pyelogram was normal.
Adrenal cortical function and reserve were normal as regards hydrocortisone and 17-ketosteroid excretion. Aldosterone excretion was elevated, as shown in numerous metabolism studies ( vide infra).
Urinary concentrating ability was somewhat impaired: dehydration for twenty-four hours induced a loss of 2.5 per cent of body weight and a rise of urine osmolality to 717 mOsm per kg., not further increased by the administration of Pitressin? (aqueous Pitressin 10 units intramuscularly). Glomerular filtration rate (inulin) was 30 ml. per minute or 68 ml. per minute per 1.73 M2 . Twenty-four hour urinary amino acid nitrogen was 1.10 per cent of total nitrogen (normal, 1 to 2 per cent).
The patient had always been a poor eater and craved salt. He was constipated and required an enema every other day. Prior to correction of the hypokalemia he had several convulsive seizures which did not resemble carpopedal spasm; these did not recur while the hypokalemia was being successfully treated. Before operation this was accomplished with potassium chloride supplements (134 mEq. per day), a low sodium diet and human serum albumin given intravenously once a week. Appetite and growth rate increased with treatment and constipation disappeared. The electrocardiogram reverted to normal.
In view of the persistent aldosteronism, partial adrenalectomy and renal biopsy were performed when the patient was eight years old. The findings will be discussed subsequently. Muscle analysis showed a decrease of intracellular potassium and an increase of intracellular sodium. For the next fifteen months the patient remained symptom-free without sodium restriction or potassium supplements, but serum potassium remained low (2.3 to 2.8 mEq. per L.), and urinary aldosterone high (e.g., 38 mug. per day on a daily sodium intake of 110 mEq.).
Fifteen months after surgery, tetany returned, and the patient was readmitted. Serum potassium was 2.0 mEq. per L., serum carbon dioxide 29 mEq. per L., and serum sodium 139 mEq. per L. Urinary aldosterone was 5 mug. per day, and aldosterone secretion 750 mug. a day. An assay of serum angiotensin concentration gave a figure of 240 mmug. per 100 ml., an approximately eightfold increase above the normal figure.
Patient M. W. was a twenty-five year old Negro man with a history of enuresis, slow growth, weakness and fatigue which had prevented him from entering school until the age of twelve. He had been seen first at twelve years of age in the Pediatric Clinic, Duke University Medical Center, with a history of vomiting and pain in the abdomen and calves. At the age of sixteen he had been admitted to Duke University Medical Center in a semicomatose condition with a blood pressure of 104/70 mm. Hg. Chemical analysis revealed the following serum values: potassium 1.28 mEq. per L., chloride 41.3 mEq. per L., and carbon dioxide 58.6 mEq. per L. Proteinuria was present. The electrocardiogram was reported as showing changes of hypokalemia. Pyelography revealed dilation of the middle third of both ureters. The patient was treated with potassium chloride and rapidly recovered.
He returned to the clinic at the age of nineteen. He had not taken potassium chloride consistently, and weakness, polydipsia, polyuria, enuresis and episodic cramps in his hands and legs had persisted. Serum potassium and chloride were 1.8 and 82 mEq. per L., respectively. He was admitted to Duke University Medical Center for study. Chvostek’s and Trousseau’s signs were elicited. The urine again contained protein (2 plus). He excreted 65 per cent of a dose of phenolsulfonphthalein in two hours. Serum potassium and carbon dioxide were 2.0 and 30 mEq. per L., respectively, non-protein nitrogen was 30 mg. per 100 ml. Serum calcium, phosphorus, total protein and albumin and globulin were normal. Eosinophil count was 55 per cu. mm., with a fall to 5 per cu. mm. after an eight-hour infusion of ACTH. He was then referred to the National Institutes of Health for further studies.
Physical examination revealed a thin Negro boy 65 inches tall who weighed 45.8 kg. Blood pressure ranged between 124/70 and 94/52 mm. Hg on numerous determinations, with an average of 105/75 mm. Hg. Chvostek’s sign was positive. Physical and neurologic examinations were otherwise non-contributory.
Numerous urine specimens showed a pH of 7 or above, specific gravity 1.014 or below, trace to 1-plus albumin and no sugar. Chemical analysis revealed the following serum values: potassium 2.0 mEq. per L., carbon dioxide 37 mEq. per L., chloride 77 mEq. per L., sodium 132 mEq. per L., blood urea nitrogen 9 mg. per 100 ml. The result of a serologic test for syphilis was negative.
An electroencephalogram showed “diffuse dysrhythmia with no epileptiform discharges or focal abnormalities.” Repeated electrocardiograms revealed “T-vector of low magnitude, S-T segment depression and prominent U waves in leads V2 through V4 .” A roentgenogram revealed that the epiphyses of the radius, ulna and iliac crest had not fused; calcific deposits were present in the region of the calyces in both kidneys. Urinary excretion of 17-hydroxycorticoids and 17-ketosteroids was 1.8 and 5.6 mg. per day, respectively. Aldosterone excretion was elevated ( vide infra).
Maximal urinary osmolality after dehydration for four hours and Pitressin (aqueous Pitressin 200 milliunits intravenously and 500 milliunits subcutaneously every thirty minutes) was only 254 mOsm per kg. Glomerular filtration rate (creatinine) was 105 ml. per minute. He has consistently refused adrenal exploration; tissue was obtained from the right kidney by percutaneous punch biopsy. An assay of serum angiotensin concentration gave a figure of about 70 mmug. per 100 ml., a clear increase above the normal figure.
Metabolic balance studies were carried out to measure (1) the effects of changing sodium intake (see Fig. 1 , 2 and 6 ), (2) the effects of changing potassium intake ( Fig. 2 and 3 ), (3) the effects of acid, (4) the effects of albumin given intravenously ( Fig. 3 and 11 ), and (5) the effects of spirolactones ( Fig. 4 and 5 ). (These studies were performed prior to surgery in C.J.). The final studies were made to determine  the effects of angiotensin and of renin (after surgery in C. J.). The adrenal (C.J.) and renal (C.J. and M. W.) biopsy specimens were studied by several histochemical technics ( Fig. 9 through 14 ). Many of the methods employed have been previously reported from this laboratory  . In addition, in some studies urinary aldosterone was determined by the method of Kliman and Peterson  , urinary titratable-acidity-minus-bicarbonate was measured by a modification of the method of Dawson and associates  , urinary ammonium by the method of Conway  , urinary 17-ketosteroids by a modification of the method of Zimmerman  , and urinary and plasma 17-hydroxy-corticoids by the method of Peterson  .
On a high sodium intake (67 and 117 mEq. in C. J. and M. W., respectively) both patients showed hypokalemic alkalosis and excreted excessive amounts of aldosterone (20 to 40 mug. per day). When the sodium intake was lowered to 17 mEq. a day, urinary sodium fell to intake values in C.J. (Fig. 1) serum potassium and carbon dioxide returned towards normal, and serum sodium did not change. Urinary potassium did not change.
Figure 1. Patient C.J. Effect of changing sodium intake on serum sodium, potassium and carbon dioxide concentration, sodium and potassium balance, urinary hydrogen ions and aldosterone and body weight.
Figure 2. Patient M. W. Effect of changing sodium intake on serum sodium, potassium and carbon dioxide concentration, urinary sodium, potassium, hydrogen ions and aldosterone and body weight.
Figure 3. Patient C.J. Effect of stepwise reduction of potassium intake on serum potassium and carbon dioxide concentration, sodium and potassium balance and body weight.
In another study (Fig. 6) , the sodium intake was abruptly increased from 6 to 106 mEq. per day. This induced loss of potassium, a fall of serum potassium from 5.1 to 2.0 mEq. per L. and a rise of serum carbon dioxide from 25 to 34 mEq. per L.
In M. W., on the other hand (Fig. 2) , when sodium intake was lowered to 17 mEq. per day, urinary sodium exceeded the sodium intake, serum potassium did not change, serum carbon dioxide rose and serum sodium fell. Urinary potassium did not change. After four days on this regimen (Fig. 2) symptoms of hypokalemia became worse, and potassium intake was increased as a therapeutic measure.
In C.J., when potassium intake was lowered stepwise (Fig. 3) , potassium balance became slightly negative, urinary potassium not falling significantly
Figure 5. Patient M. W. Effect of human serum albumin infusion on serum sodium, potassium and carbon dioxide concentration, blood hemoglobin and hematocrit, sodium and potassium balances, urinary aldosterone and body weight.
below intake despite restriction of sodium intake to 6 mEq. per day, and serum potassium fell while serum carbon dioxide rose. In another study (not shown) when potassium intake was lowered abruptly, the same results were obtained and urinary aldosterone fell progressively–presumably an effect of potassium depletion per se  .
In M. W., on an intake of 70 mEq. of potassium a day (Fig. 2) , urinary potassium remained close to the potassium intake despite restriction of sodium intake to 17 mEq. per day, serum potassium remained below 2 mEq. per L. and serum carbon dioxide rose. As already noted, the patient’s clinical condition precluded a study with a lower potassium intake.
Effects of Albumin Given Intravenously ( Fig 5 ).
When human serum albumin was given intravenously, both patients showed large decreases in hemoglobin concentration and hematocrit and increases in body weight. Urinary sodium fell from 2 (C.J.) and 36 (M.W.) mEq. per day, to zero in both patients. In C.J., who received 5 mEq. of sodium a day, serum sodium fell; while in M. W., who received 60 mEq. of sodium a day, it rose. In both patients there was retention of potassium and a rise of serum potassium. Urinary aldosterone rose in C.J., and did not change in M. W.
When aldosterone antagonists were given there was an increase in urinary sodium, with a negative sodium balance, in both patients. In both patients there was a fall of urinary potassium with potassium retention, as serum potassium rose sharply (from 2.6 to 6.8 mEq. per L.) in C.J. and slight (from 2.2 to 2.6 mEq. per L.) in M. W. Serum carbon dioxide had decreased by the first day after treatment in C. J., whereas it decreased throughout treatment in M. W. Urinary aldosterone rose in both patients.
Effects of Renin and of Angiotensin.
During the studies with renin and angiotensin serum potassium ranged from 2.4 to 3.8 mEq. per L. in C. J., and from 1.6 to 2.4 mEq. per L. in M. W.
When angiotensin II was given to C. J. intravenously at rates of 0.005, 0.01, 0.05 and 0.1 mug. per kg. per minute (weight 17.7 kg.), blood pressure increased from control levels (range: 90/55 to 96/60 mm. Hg) to sustained peak values of 92/64, 98/60, 106/80 and 128/98 mm. Hg for each of these doses, respectively; the pulse rate decreased from control levels (range: 80 to 88) by 6 to 12 beats per minute with each dose.
When angiotensin II was given to M. W. intravenously at rates of 0.05, 0.1, 0.2 and 0.4 mug. per kg. per minute (weight
Figure 6. Patient C. J. Effect of SC-9420 (Aldactonc) and a low sodium intake on serum potassium and carbon dioxide concentration, urinary sodium, potassium and aldosterone and body weight. Also the effect of high sodium intake on these same measurements.
45.8 kg.), blood pressure increased from control levels (range: 104/66 to 114/76 mm. Hg) to peak values of 118/84, 124/88, 140/104 and 172/112 mm. Hg for each of these doses, respectively; the pulse rate decreased from control levels (range: 72 to 78) by 4 to 8 beats per minute with each dose. Blood pressure then declined from these peak values, despite continued infusion of angiotensin, to pressures of 110/80, 112/82, 128/94 and 140/108 mm. Hg, respectively, with the doses given. When angiotensin I was infused at the rate of 5 mug. per minute the blood pressure increased from 106/70 to 114/80 mm. Hg. The effects of angiotensin II on blood pressure in these patients and in several normal subjects are shown graphically in Figure 8 .
Patient C. J.
The kidneys and adrenal glands in C. J. appeared to be normal upon gross examination at the time of operation. The left adrenal gland was excised in toto and weighed 2.7 gm. Eighttenths of a gram of tissue was removed from the right adrenal gland, and a biopsy of the right kidney was performed.
Figure 7. Patient M. W. Effect of SC-9420 (Aldactone) on serum sodium, potassium and carbon dioxide concentration, sodium and potassium balances, urinary hydrogen ions and aldosterone and body weight. There were no fecal measurements during the last five days.
Adrenal tissue was fixed in 4 per cent formaldehyde, embedded and cut in the usual way, and colored with hematoxylin and eosin. Histologically, the sections showed limited areas of hypertrophy of the zona glomerulosa which consisted of two distinct layers: (1) a compact outer layer of small cells with hyperchromatic nuclei, and (2) an inner layer of large, clear cells arranged in the form of blind tubules, abutting the peripheral cells, which did not seem to be a part of the zona fasciculata. The thickness of the zona glomerulosa, considered as consisting of both of these glomerulosa layers, was approximately twice that of the zona glomerulosa of a normal child of the same age. In some areas in which the peripheral layer was hyperplastic, the inner tubular layer was not discernible.
The kidney specimen consisted of a wedge about 10 mm. wide and 4 mm. thick. It contained about 60 glomeruli. It was cut in two, and one section was fixed for six hours in Zenker’s formol solution (90 ml. of stock Zenker fluid and 10 ml. of 40 per cent formaldehyde). The other section was fixed and refrigerated for four days in Regaud’s fluid (80 ml. of 3 per cent potassium bichromate and 20 ml. of 40 per cent formaldehyde) and mordanted in a refrigerator for eight days in 3 per cent potassium bichromate for mitochondrial stains  . Sections were cut 5 and 6 mu thick and colored by ordinary hematoxylin and eosin, the Cowdry-Bensley modification of Altmann’s acid-fuchsin-aniline blue, the periodic acid-Schiff reaction, or by Bowie’s strain   for granules of the juxtaglomerular cells.
About 60 per cent of the uriniferous tubules showed hyperplasia of the juxtaglomerular apparatus, and the walls of the glomerular vessels of about 20 per cent of these were hyperplastic ( Fig. 9 and 10 ). In these hyperplastic apparatuses the wall of the afferent arteriole was found to be 80 to 100 mu in thickness, whereas the wall of this vessel in the kidney of a normal child of the same age was only 30 mu in thickness. The juxtaglomerular cells of only two Malpighian bodies contained so-called JG granules, colorable with Bowie’s stain. The cells of the juxtaglomerular perivascular collar (juxtaglomerular cells of Hartroft) were found to be long, thin and fibroblastic in nature, unlike the glandular type of cell in the normal apparatus.
The macula densa of the uriniferous tubule was enlarged to accommodate as many as 30 or 40 nuclei within the mural roof adjacent to the juxtaglomerular cells and the prominent pohlkissen body. The basement membrane was continuous throughout the length of the afferent arteriole, forming an abnormal membrane obstructing the space normally existing between the juxtaglomerular wall and the macula densa. (There is no such membrane in this position in a normal kidney).
Print This Post