Use of Human Growth Hormone Combined with Nutritional Critical Care Unit THOMAS R.
From the
Support in a
ZIEGLER, M.D., LORRAINE S. YOUNG, R.D., M.S., ELISABETTA FERRARI-BALIVIERA, M.D., ROBERT H. DEMLING, M.D., AND DOUGLAS W. WILMORE, M.D.
Laboratory of Surgical Metabolism and Nutrition and the Department of Surgery, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts
ABSTRACT. The administration of growth factors may potentially accelerate recovery during critical illness by reducing body protein catabolism, enhancing wound healing, and improving skeletal muscle function. The purpose of this phase 1 study was to evaluate the safety and initial efficacy of a recombinant growth factor, human growth hormone (GH), combined with nutritional support in a critical care unit. Following an initial control week, 11 individuals received GH (10 mg/day) daily for 1-6 consecutive weeks. Near constant nutrient intake was provided via parenteral and/or enteral feedings throughout the study period. Vital signs and other clinical parameters, blood values, and nutrient excretion were monitored daily. GH administration was not associated with clinically significant adverse effects. During the first 2 weeks of study, nitrogen excretion decreased from 1356 ± 157 mmol/day (19.0 ± 2.2 g/
Erosion of lean
body
mass
characteristically
occurs
following stresses such as trauma, infection, and elective operation. In severely injured patients, specialized nutritional support is commonly administered in an attempt to attenuate the negative nitrogen balance which occurs; however, rarely, if ever, is body protein increased or even maintained in patients following significant catabolic insults. Catabolic states are associated with immunosupression,2 impaired wound healing and delayed tissue repair,’ respiratory muscle weakness,’ and decreased ambulation because of impaired skeletal muscle function5; thus, loss of body protein is interrelated with the complications and prolonged convalescence in critically ill patients. The current availability of genetically engineered growth factors for clinical investigation provides new opportunities to study their effect on nutritional support efficacy and subsequent clinical outcomes.’ Over the past 5 years, we have studied the effects of recombinant human growth hormone (GH) on protein metabolism in normal volunteers, 7,8 patients with gastrointestinal disease requiring parenteral nutrition,’ and in stable postoperative patients.’o Administration of GH has increased protein synthesis and improved nitrogen balance by 2-4 g/day (143-246 mmol/day). This anabolic response occurred even when only one half of the necessary calories were provided. While the combination of growth hor-
Reprint requests: Douglas W. Wilmore, Laboratory for Surgical Metabolism and Nutrition, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02215.
day) during control to 899 ± 107 mmol/day (12.6 ± 1.4 g/day) with growth hormone (p < 0.002) in association with markedly reduced urea generation. Significant reductions in potassium excretion (control 100 ± 11 mmol/day us 69 ± 6 with GH; p < 0.01) and phosphorus excretion (31 ± 5 mmol/day us 18 ± 3; p < 0.025) also occurred during GH. The protein-conserving effects of GH were sustained during several weeks of treatment. Growth hormone enhanced the efficiency of administered protein and facilitated nitrogen retention without clinically significant adverse effects in this small patient group. Controlled trials are indicated to determine whether use of this anabolic hormone reduces hospitalization time and improves other clinical outcomes in severely injured patients when combined with appropriate nutritional support. (Journal of Parenteral and Enteral Nutrition 14:574-581, 1990)
mone
and hypocaloric
gen balance for 3-4 viduals and not in
feeding resulted in positive nitroweeks, this occurred in stable indipatients following major catabolic
insults.9 Studies in the 1960s and 1970s demonstrated that short courses (~l week) of pituitary-derived GH promoted protein retention following severe burn injury.’’13 However, the metabolic effects of recombinant GH in severely catabolic patients when administered for longer periods or the safety of this approach has not been documented. Therefore, the purpose of this phase 1 study was to evaluate whether improved nutritional efficacy could be safely achieved with GH administration in burn/ trauma unit patients. This investigation demonstrates that recombinant GH given daily for several weeks safely enhances protein retention throughout the postinjury period in patients requiring intensive care and nutrition
support. METHODS
Patients The study was performed in the burn/trauma unit of Brigham and Women’s Hospital (Boston, MA). Eleven injured individuals (average age of 34 years, range 1883 ; Table I) were studied during a stable course of their illness following severe burn or vehicular trauma. Investigations were performed in six patients during the early phase of injury (7-18 days postinjury) and the other studies were done during the later convalescent phase (23-76 days postinjury). No patient had clinical evidence or history of malignancy, diabetes, or other endocrinoa
574 Downloaded from pen.sagepub.com at University of Sussex Library on August 14, 2015
575 TABLE I Clinical characteristics
BSA, body surface area; MVA, motor vehicle accident; ARDS, Required mechanical ventilation at entry into the study.
of patients
adult respiratory distress
syndrome.
*
pathies, and none received corticosteroids prior to or during the study period. The research protocol was approved by the Committee for the Protection of Human Subjects from Research Risk of Brigham and Women’s Hospital in accordance with the principles for human experimentation defined in the declaration of Helsinki. Informed consent obtained from all subjects or their guardians.
was
Study Design Patients were entered into the study after a stable and adequate nutrient intake had been achieved for at least 5 days. Energy requirements for each patient were determined using previously described methodology.&dquo; The exception was subject 10, whose calorie prescription was based on requirements appropriate for spinal cord-injured patients.&dquo; Protein intake generally provided 1.2-1.5 g of protein/ kg/day, and calories were provided at 30-35 kcal/kg/day. Nutrients
administered via intravenous catheters as indicated, and ad libitum oral food was allowed as tolerated. Carbohydrate represented the major energy source (60-70% of nonprotein calories) and fat provided the remaining nonprotein calories. Adequate vitamin, mineral, and trace metal requirements were also administered. The study was initiated by the administration of the were
and/or feeding tubes
prescribed diet combined with continuous 24-hr collections of all urine, stool, emesis, and tube drainage output between 0600-0600 hr. Wound exudate was not collected. Following an initial control week, the diet and collections were continued for 1-6 additional weeks. During this period, recombinant methionyl human GH (10 mg; Protropin, Genentech Inc, South San Francisco, CA) was administered at 0830 hr via subcutaneous injection. When expressed per kilogram of body weight, GH dose ranged from 0.096-0.172 mg/kg/day and averaged 0.131 ± 0.007 mg/kg/day.
Clinical and Metabolic Measurements
Laboratory data, respiratory function measurements, and vital signs were determined as clinically indicated. Additional blood samples were obtained 1-3 times weekly between 0700 and 0830 hr for determination of GH, insulin-like growth factor I (IGF 1), insulin, free fatty acid, and anti-human GH antibodies. Determination of daily fluid and nutrient intake and excretion and the analysis of blood concentrations were performed as previously described.’ Calculations The mean daily value for all vital signs (0600-0600 hr) calculated for each patient, and available body
was
Downloaded from pen.sagepub.com at University of Sussex Library on August 14, 2015
576
weights were averaged for each study week. Because nitrogen and mineral exudate from the burned surface and other wounds was not measured, true metabolic balances for these compounds was not determined. Thus, the measured daily loss (urine and other measured losses) used to indicate relative elimination and retention of individual nutrients. Data from the initial 24 hr of the control and first treatment week were considered equilibration periods and were not included in the calculation of weekly mean data. Data obtained during all 7 days of each subsequent study week were used to calculate weekly means. Average blood values obtained during the last 4 days of the control and initial GH treatment week in the 11 patients were compared. For the seven patients studied for at least 28 days (patients 4, 5, 6, 7, 8, 10, and 11), the mean blood value for each consecutive 4-day period was calculated. Serum calcium was adjusted for the corresponding serum albumin concentration by standard tech-
was
rates
niques.16 Statistics Statistical calculations
were
performed
on a
personal
computer (Macintosh SE, Apple Computer Inc, Cuper-
tino, CA) using a statistical software package (Statview 512+, Brainpower Inc, Calabasas, CA) and on an IBM4341 computer with MINITAB (Pennsylvania State University, State College, PA). Paired t-testing, re-
peated-measures analysis of variance (ANOVA), and linear regression techniques were used as appropriate. A significant difference between means was accepted when the p value was
From the
Support in a
ZIEGLER, M.D., LORRAINE S. YOUNG, R.D., M.S., ELISABETTA FERRARI-BALIVIERA, M.D., ROBERT H. DEMLING, M.D., AND DOUGLAS W. WILMORE, M.D.
Laboratory of Surgical Metabolism and Nutrition and the Department of Surgery, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts
ABSTRACT. The administration of growth factors may potentially accelerate recovery during critical illness by reducing body protein catabolism, enhancing wound healing, and improving skeletal muscle function. The purpose of this phase 1 study was to evaluate the safety and initial efficacy of a recombinant growth factor, human growth hormone (GH), combined with nutritional support in a critical care unit. Following an initial control week, 11 individuals received GH (10 mg/day) daily for 1-6 consecutive weeks. Near constant nutrient intake was provided via parenteral and/or enteral feedings throughout the study period. Vital signs and other clinical parameters, blood values, and nutrient excretion were monitored daily. GH administration was not associated with clinically significant adverse effects. During the first 2 weeks of study, nitrogen excretion decreased from 1356 ± 157 mmol/day (19.0 ± 2.2 g/
Erosion of lean
body
mass
characteristically
occurs
following stresses such as trauma, infection, and elective operation. In severely injured patients, specialized nutritional support is commonly administered in an attempt to attenuate the negative nitrogen balance which occurs; however, rarely, if ever, is body protein increased or even maintained in patients following significant catabolic insults. Catabolic states are associated with immunosupression,2 impaired wound healing and delayed tissue repair,’ respiratory muscle weakness,’ and decreased ambulation because of impaired skeletal muscle function5; thus, loss of body protein is interrelated with the complications and prolonged convalescence in critically ill patients. The current availability of genetically engineered growth factors for clinical investigation provides new opportunities to study their effect on nutritional support efficacy and subsequent clinical outcomes.’ Over the past 5 years, we have studied the effects of recombinant human growth hormone (GH) on protein metabolism in normal volunteers, 7,8 patients with gastrointestinal disease requiring parenteral nutrition,’ and in stable postoperative patients.’o Administration of GH has increased protein synthesis and improved nitrogen balance by 2-4 g/day (143-246 mmol/day). This anabolic response occurred even when only one half of the necessary calories were provided. While the combination of growth hor-
Reprint requests: Douglas W. Wilmore, Laboratory for Surgical Metabolism and Nutrition, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02215.
day) during control to 899 ± 107 mmol/day (12.6 ± 1.4 g/day) with growth hormone (p < 0.002) in association with markedly reduced urea generation. Significant reductions in potassium excretion (control 100 ± 11 mmol/day us 69 ± 6 with GH; p < 0.01) and phosphorus excretion (31 ± 5 mmol/day us 18 ± 3; p < 0.025) also occurred during GH. The protein-conserving effects of GH were sustained during several weeks of treatment. Growth hormone enhanced the efficiency of administered protein and facilitated nitrogen retention without clinically significant adverse effects in this small patient group. Controlled trials are indicated to determine whether use of this anabolic hormone reduces hospitalization time and improves other clinical outcomes in severely injured patients when combined with appropriate nutritional support. (Journal of Parenteral and Enteral Nutrition 14:574-581, 1990)
mone
and hypocaloric
gen balance for 3-4 viduals and not in
feeding resulted in positive nitroweeks, this occurred in stable indipatients following major catabolic
insults.9 Studies in the 1960s and 1970s demonstrated that short courses (~l week) of pituitary-derived GH promoted protein retention following severe burn injury.’’13 However, the metabolic effects of recombinant GH in severely catabolic patients when administered for longer periods or the safety of this approach has not been documented. Therefore, the purpose of this phase 1 study was to evaluate whether improved nutritional efficacy could be safely achieved with GH administration in burn/ trauma unit patients. This investigation demonstrates that recombinant GH given daily for several weeks safely enhances protein retention throughout the postinjury period in patients requiring intensive care and nutrition
support. METHODS
Patients The study was performed in the burn/trauma unit of Brigham and Women’s Hospital (Boston, MA). Eleven injured individuals (average age of 34 years, range 1883 ; Table I) were studied during a stable course of their illness following severe burn or vehicular trauma. Investigations were performed in six patients during the early phase of injury (7-18 days postinjury) and the other studies were done during the later convalescent phase (23-76 days postinjury). No patient had clinical evidence or history of malignancy, diabetes, or other endocrinoa
574 Downloaded from pen.sagepub.com at University of Sussex Library on August 14, 2015
575 TABLE I Clinical characteristics
BSA, body surface area; MVA, motor vehicle accident; ARDS, Required mechanical ventilation at entry into the study.
of patients
adult respiratory distress
syndrome.
*
pathies, and none received corticosteroids prior to or during the study period. The research protocol was approved by the Committee for the Protection of Human Subjects from Research Risk of Brigham and Women’s Hospital in accordance with the principles for human experimentation defined in the declaration of Helsinki. Informed consent obtained from all subjects or their guardians.
was
Study Design Patients were entered into the study after a stable and adequate nutrient intake had been achieved for at least 5 days. Energy requirements for each patient were determined using previously described methodology.&dquo; The exception was subject 10, whose calorie prescription was based on requirements appropriate for spinal cord-injured patients.&dquo; Protein intake generally provided 1.2-1.5 g of protein/ kg/day, and calories were provided at 30-35 kcal/kg/day. Nutrients
administered via intravenous catheters as indicated, and ad libitum oral food was allowed as tolerated. Carbohydrate represented the major energy source (60-70% of nonprotein calories) and fat provided the remaining nonprotein calories. Adequate vitamin, mineral, and trace metal requirements were also administered. The study was initiated by the administration of the were
and/or feeding tubes
prescribed diet combined with continuous 24-hr collections of all urine, stool, emesis, and tube drainage output between 0600-0600 hr. Wound exudate was not collected. Following an initial control week, the diet and collections were continued for 1-6 additional weeks. During this period, recombinant methionyl human GH (10 mg; Protropin, Genentech Inc, South San Francisco, CA) was administered at 0830 hr via subcutaneous injection. When expressed per kilogram of body weight, GH dose ranged from 0.096-0.172 mg/kg/day and averaged 0.131 ± 0.007 mg/kg/day.
Clinical and Metabolic Measurements
Laboratory data, respiratory function measurements, and vital signs were determined as clinically indicated. Additional blood samples were obtained 1-3 times weekly between 0700 and 0830 hr for determination of GH, insulin-like growth factor I (IGF 1), insulin, free fatty acid, and anti-human GH antibodies. Determination of daily fluid and nutrient intake and excretion and the analysis of blood concentrations were performed as previously described.’ Calculations The mean daily value for all vital signs (0600-0600 hr) calculated for each patient, and available body
was
Downloaded from pen.sagepub.com at University of Sussex Library on August 14, 2015
576
weights were averaged for each study week. Because nitrogen and mineral exudate from the burned surface and other wounds was not measured, true metabolic balances for these compounds was not determined. Thus, the measured daily loss (urine and other measured losses) used to indicate relative elimination and retention of individual nutrients. Data from the initial 24 hr of the control and first treatment week were considered equilibration periods and were not included in the calculation of weekly mean data. Data obtained during all 7 days of each subsequent study week were used to calculate weekly means. Average blood values obtained during the last 4 days of the control and initial GH treatment week in the 11 patients were compared. For the seven patients studied for at least 28 days (patients 4, 5, 6, 7, 8, 10, and 11), the mean blood value for each consecutive 4-day period was calculated. Serum calcium was adjusted for the corresponding serum albumin concentration by standard tech-
was
rates
niques.16 Statistics Statistical calculations
were
performed
on a
personal
computer (Macintosh SE, Apple Computer Inc, Cuper-
tino, CA) using a statistical software package (Statview 512+, Brainpower Inc, Calabasas, CA) and on an IBM4341 computer with MINITAB (Pennsylvania State University, State College, PA). Paired t-testing, re-
peated-measures analysis of variance (ANOVA), and linear regression techniques were used as appropriate. A significant difference between means was accepted when the p value was
