Intrauterine growth restriction (IUGR) is associated with accelerated growth after birth. Together, IUGR and accelerated growth after birth predict reduced lean tissue mass and increased obesity in later life. Although placental insufficiency is a major cause of IUGR, whether it alters growth and adiposity in early postnatal life is not known. We hypothesized that placental restriction (PR) in the sheep would reduce size at birth and increase postnatal growth rate, fat mass, and feeding activity in the young lamb. PR reduced survival rate and size at birth, with soft tissues reduced to a greater extent than skeletal tissues and relative sparing of head width (P < 0.05 for all). PR did not alter absolute growth rates (i.e., the slope of the line of best fit for age vs. parameter size from birth to 45 days of age) but increased neonatal fractional growth rates (absolute growth rate relative to size at birth) for body weight (+24%), tibia (+15%) and metatarsal (+18%) lengths, hindlimb (+23%) and abdominal (+19%) circumferences, and fractional growth rates for current weight (P < 0.05) weekly throughout the first 45 days of life. PR and small size at birth reduced individual skeletal muscle weights and increased visceral adiposity in absolute and relative terms. PR also altered feeding activity, which increased with decreasing size at birth and was predictive of increased postnatal growth and adiposity. In conclusion, PR reduced size at birth and induced catch-up growth postnatally, normalizing weight and length but increasing adiposity in early postnatal life. Increased feeding activity may contribute to these alterations in growth and body composition following prenatal restraint and, if they persist, may lead to adverse metabolic and cardiovascular outcomes in later life.
- fetal growth restriction
- catch-up growth
intrauterine growth restriction (IUGR) in humans, evident as reduced weight and length and/or increased thinness for gestational age, is associated with altered postnatal growth in the first few months of life (10, 33, 48). Thus, in many IUGR infants, acceleration of growth for their size in the first few months of life, termed catch-up growth (4, 31, 33, 34, 36, 37, 48), along with their small size at birth, predicts impairment of later function and health, including obesity (8, 45).
This association between catch-up growth in humans and increased subcutaneous and visceral obesity in later life (51) could partly account for the concomitantly increased risk of developing adult-onset diseases, such as diabetes, hypertension, and cardiovascular disease (11, 20, 23, 49). It is unclear, however, whether the increased adiposity associated with catch-up growth following IUGR is established early in postnatal life at the time of greatest catch-up and before onset of these other conditions or emerges later.
Low-birth-weight lambs born to ewes carrying triplets also grow more quickly for their size after birth and have a greater percentage of body fat at 20 kg live weight than do high-birth-weight lambs (27), although their adiposity has not been compared at the same age. In addition, the increased adiposity of low-birth-weight lambs was further exacerbated with ad libitum feeding compared with restricted feeding, suggesting that altered appetite and/or satiety may contribute to accelerated growth and obesity (27).
Placental size accounts for much of the variation in size at birth in mammalian species (44, 47), and placental insufficiency is a major cause of IUGR, but whether it also induces catch-up growth in early postnatal life has not been determined, and its role in later outcomes is unclear. Experimental placental restriction in late gestation in the rat induces diabetes, hypertension, and obesity in adult offspring but generally does not induce accelerated or catch-up growth in the first few weeks of life before weaning or puberty (5, 43, 52). This might in part reflect the relative immaturity of the rat at birth or its polytocous nature, where placental restraint of mammary gland growth and function may continue to limit growth postnatally, especially with many offspring. Similarly, placental restriction imposed in late gestation in the sheep, by umbilicoplacental embolization from 120 days of gestation to full term at 146 days (14), reduces size at birth to ∼50% of controls but does not induce neonatal catch-up growth (30). One factor contributing to the lack of catch-up growth in these studies (5, 14, 30, 58) may have been the relatively late stage of gestation and fetal development at which placental restraint was imposed compared with the earlier onset of much human placental insufficiency (6). We previously showed that restriction of implantation and placental growth in the sheep limits delivery of oxygen and nutrients to the fetus (40–42, 55), which alters its growth from midgestation onward. Recently, we also showed that placental restriction in the sheep reduces size at birth, increases neonatal growth rates, and, in a limited number of lambs, increases adiposity (16). Also, placental restriction and small size at birth reduce plasma total thyroxine and increase plasma total triiodothyronine concentrations postnatally, whereas catch-up growth relates to the increased abundance of the more bioactive forms of thyroid hormone (17). We therefore hypothesized that restriction of placental implantation and growth and, hence, earlier onset of restriction of fetal growth would reduce size at birth and increase postnatal growth rate and fat mass in the young lamb. We also hypothesized that placental restriction would increase feeding activity in the young lamb and that this would be predictive of their postnatal growth rate and adiposity.
MATERIALS AND METHODS
Animals and surgery.
All procedures were approved by the Adelaide University Animal Ethics Committee. Placental and fetal size was restricted in 60 Merino ewes by removal of the majority of visible endometrial caruncles from the nonpregnant uterus, leaving three to four caruncles in each horn of the bicornuate uterus (50). This reduces the number of placentomes formed in the subsequent pregnancy and, hence, placental size and function (50). Surgery was performed under aseptic conditions, and general anesthesia was induced by injection of thiopental sodium (9.9–14 mg/kg iv; Pentothal, Rhone Merieux) and maintained with halothane inhalation anesthetic (Zeneca) in oxygen. After surgery, all ewes received a 3-day course of intramuscular antibiotics (2 ml; Ilium Penstrep, Troy Laboratories). After ≥6 wk of recovery, the ewes underwent dated matings, and pregnancies were confirmed by ultrasound. From ∼1 wk before predicted full-term birth, the ewes were housed in individual pens in animal holding rooms with a 12:12-h light-dark cycle and were fed lucerne chaff twice daily ad libitum. Water was available to ewes and lambs ad libitum, except during measurement of feeding activity (see below). These ewes delivered 50 lambs: 27 (17 male and 10 female) from control ewes and 23 (9 male and 14 female) from placentally restricted ewes. Neonatal growth and adiposity data from a subset of these lambs (9 control and 9 placentally restricted) have been presented previously in a report of effects of placental restriction on thyroid hormone abundance and the relations between circulating thyroid hormones and growth (17). The placenta was collected after delivery and weighed where possible. Eight pairs of twins (5 control and 3 placentally restricted) were included for analyses of perinatal survival only. The lambs were housed in the pens with their mothers throughout the study and were not weaned.
At birth and at 5-day intervals up to 45 days of age, weight, crown-rump length (CRL), tibia length, radius/ulna length, metatarsal and metacarpal lengths, shoulder height, skull width and length, and abdominal, thoracic, radius/ulna, and hindlimb (around the knee and midway around the tibia) circumferences were measured. Body mass index was calculated as weight/CRL2 (kg/cm2), and ponderal index was calculated as weight/CRL3 (kg/cm3). The absolute growth rate (AGR) for all parameters was linear for 45 ± 3 days after birth and was the slope of the line of best fit as determined by linear regression analysis for each parameter within an individual sheep. Neonatal fractional growth rate (NFGR) was calculated as the AGR from 0 to 45 days of age for a parameter divided by the size of that parameter at birth. Current fractional growth rates (CFGR) at weekly intervals were calculated as the absolute growth rate for 0–45 days divided by the value for a parameter at the start of that week.
Feeding activity was characterized in a subset of singleton lambs (9 control and 6 placentally restricted) at 15 ± 1 days of age. Briefly, lambs were fasted for 1 h (isolated from ewe) and then observed with the ewe for 2.5 h for measurement of the total number of suckling events and the total suckling time. Feed and water were removed from the ewe and lamb during the 2.5-h observation period. Blood was sampled every 15 min throughout the experiment. Initially, nine control and six placentally restricted lambs were observed for the first 30 min (two 15-min intervals) after the fasting period, but three control and four placentally restricted lambs were observed for 90 min (six 15-min intervals) after cessation of fasting.
At 43 ± 2 days of age, a designated number of the singleton lambs (9 control and 6 placentally restricted) were killed by intravenous administration of an overdose of barbiturate (pentobarbitone, Lethabarb). Major organs were removed, weighed, and measured. Individual muscles (biceps, flexor carpi radialis, tibialis, semitendinosus, gastrocnemius, soleus, extensor digitorum longus, biceps femoris, and vastus lateralis) were completely removed and weighed. The summed mass of these individual muscles is termed the summed muscle mass. The fat depots that could be completely dissected were weighed (retroperitoneal, perirenal, and omental fat depots). Visceral fat was calculated as the sum of these three fat depots.
Values are means ± SE, unless otherwise stated. The effects of placental restriction and sex, as between factors, on parameters were assessed by analysis of variance (ANOVA), as were their effects, together with that of age, as a repeated-measures factor (7 levels), on CFGR. In addition, the effects of placental restriction and sex, as between factors, on individual muscle weights and adipose depot weights (absolute value or relative to body weight), as repeated-measures factors (muscle: 9 levels; fat: 3 levels), were determined by ANOVA. Where interactions between placental restriction and sex were present, post hoc tests were carried out to determine the effect of placental restriction in males and females separately. Associations between parameters were assessed by Pearson’s correlation or multiple linear regression analysis (SPSS 13.0 software package for Windows). The coefficient of variation was calculated as follows: SD ÷ mean × 100. Statistical significance was assumed at P < 0.05.
Effect of placental restriction on survival and size at birth.
Placental restriction reduced survival rate overall: 95% in control and 82% in placentally restricted lambs (P < 0.001). Survival rate was reduced to a greater extent in twins [91% in control and 61% in placentally restricted lambs (P < 0.05)] than in singletons [95% in control and 75% in placentally restricted lambs (P < 0.05)].
In singleton lambs, placental restriction reduced placental weight by 39% [0.54 ± 0.10 kg in control (n = 10) and 0.34 ± 0.08 kg in placentally restricted (n = 9) lambs (P < 0.05)] and reduced weight (−25%), hindlimb circumference (−14%), abdominal circumference (−10%), CRL (−9%), and tibia and metatarsal lengths (−6%) at birth. Body mass index (−11%) was reduced, as was skull width (−5%), in placentally restricted lambs (Table 1). Weight, CRL, skull width, shoulder height, tibia and metatarsal lengths, and abdominal circumference at birth were greater in males than in females (P < 0.05 for all; Table 1).
Effect of placental restriction on postnatal growth.
Placental restriction reduced the AGR of skull width (P < 0.05; Table 2). Placental restriction also tended to increase AGR for skull length, tibia length, hindlimb (knee joint) circumference, and radius/ulna circumference (P < 0.1 for each; Table 2). AGR for weight and both measures of hindlimb circumference were greater for males than for females (P < 0.001, P < 0.05, and P < 0.05, respectively; Table 2).
Placental restriction increased NFGR for weight, skull length, tibia length, and hindlimb (knee joint), radius/ulna, and abdominal circumferences and decreased NFGR for skull width (P < 0.05 for all; Table 3).
In placentally restricted lambs, weight, CRL, and abdominal circumference were reduced until day 30, indicating that they had “caught-up” to control lambs in terms of these parameters (Fig. 1). Tibia length of placentally restricted lambs was similar to that of controls by day 20 (Fig. 1). In contrast, skull width was reduced in placentally restricted lambs compared with controls up to and including day 45 (Fig. 1). In females, weight, shoulder height, tibia length, and abdominal circumference were reduced compared with the same parameters in males from birth to 45 days of age, but CRL was similar to that in males by day 10 and skull width was similar by day 35 (Fig. 2).
Placental restriction increased weekly CFGR for weight throughout the first 45 days of life (P < 0.05; Fig. 3). Placental restriction did not alter weekly CFGR for CRL, shoulder height, and skull width throughout the first 45 days of life but increased CFGR of CRL and decreased CFGR of skull width at day 25 (P < 0.05; Fig. 3).
Effect of placental restriction and size at birth on muscle mass.
Placental restriction tended to reduce the summed weight of dissected muscles (P < 0.1; Table 4). Skeletal muscle was reduced in females, in particular (P < 0.05), and in biceps, gastrocnemius, soleus, and biceps femoris, specifically (P < 0.05; Table 4). As a percentage live weight, placental restriction tended to reduce flexor carpi radialis muscle mass in females (P < 0.1; Table 4). Placental restriction altered gastrocnemius muscle mass as percentage of live weight differently with sex (P < 0.005) and tended to alter gastrocnemius muscle mass (P < 0.1) differently with sex, such that gastrocnemius muscle mass decreased in females (Table 4). In males, muscles (biceps, gastrocnemius, soleus, and biceps femoris) were heavier and combined muscle mass was greater than in females (P < 0.05 for all; Table 4). Biceps femoris muscle mass (corrected for live weight) correlated positively (r = 0.87, n = 7, P < 0.0025 for control; r = 0.86, n = 4, P < 0.05 for PR) with placental weight in lambs.
In control lambs, weights of biceps, gastrocnemius, soleus, and biceps femoris muscle correlated positively with size at birth in terms of weight, CRL, and shoulder height (P < 0.05), and weights of biceps and biceps femoris muscle correlated positively with abdominal circumference (P < 0.05; Table 5). In placentally restricted lambs, combined (gastrocnemius, soleus, and biceps femoris) muscle mass was positively correlated with size at birth in terms of birth weight, CRL, shoulder height, and abdominal circumference (P < 0.05 for all; Table 6). In control lambs, summed muscle mass and individual weights of gastrocnemius, soleus, and biceps femoris muscle correlated positively with AGR in terms of weight (P < 0.05 for all; Table 5). After placental restriction, the combined (gastrocnemius, biceps femoris, and vastus lateralis) muscle mass positively correlated with AGR in terms of weight and negatively with fractional growth rate in terms of weight (P < 0.05 for all; Table 6).
Effect of placental restriction and size at birth on fat mass.
Placental restriction increased the mass of individual fat depots (perirenal) and combined visceral fat mass at 45 days of age (P < 0.05; Table 4). When individual adipose tissues were considered, placental restriction increased adipose depot mass overall (P < 0.05) and increased omental fat mass in males (P < 0.05; Table 4). Placental restriction also increased visceral fat mass relative to live weight (P < 0.05) and increased total perirenal fat mass relative to live weight (P < 0.05; Table 4).
Retroperitoneal fat (corrected for live weight) correlated negatively (r = −0.67, n = 7, P < 0.05 for control; r = −0.89, n = 4, P < 0.025 for placental restriction) with placental weight in lambs. In placentally restricted lambs, omental fat mass correlated positively with placental weight and absolute growth in terms of weight (P < 0.05 for both; Table 6). Total omental and visceral fat mass was greater in males than in females at 45 days of age (P < 0.05; Table 4). After placental restriction, visceral fat mass correlated positively with size at birth in terms of shoulder height and abdominal circumference (P < 0.05 for both; Table 6), whereas omental fat mass correlated positively with weight, shoulder height, and abdominal circumference at birth (P < 0.05 for all; Table 6).
Feeding activity and weight gain after fasting.
Weight gain from 30 to 90 min and from 30 to 150 min after the cessation of fasting correlated positively with the total suckling time (during the first 30 min after fasting: r = 0.88, n = 14, P < 0.001; r = 0.77, n = 7, P = 0.022, respectively) and total suckling time (during the first 90 min after fasting: r = 0.71, n = 7, P = 0.038; r = 0.87, n = 7, P = 0.005, respectively) in all animals. The weight gain from 30 to 90 min after fasting correlated positively with the total number of suckling events (during the first 30 min after fasting: r = 0.63, n = 14, P = 0.008) in all animals. Weight gain from 30 to 150 min after the cessation of fasting similarly correlated positively with the total number of suckling events (during the first 90 min after fasting: r = 0.81, n = 7, P = 0.014) in all animals.
Effect of placental restriction and size at birth on feeding activity.
The number of suckling events during the first 90 min after fasting varied with time (quadratic: P = 0.038) and was unaltered by placental restriction (Fig. 4A). Total suckling time during the first 90 min varied with time (quadratic: P = 0.015) and was altered by placental restriction (time * treatment cubic: P = 0.037), such that the episodic increases appeared earlier and were more sustained (Fig. 4B). The mean suckling time varied episodically during the first 90 min after fasting (cubic: P = 0.027). When birth weight was included as a covariate, the mean suckling time was altered, such that the periodic increases occurred earlier and/or were more sustained with reduced birth weight (time * birth weight cubic: P = 0.027) during the first 90 min after fasting (Fig. 4C). The number of suckling events, the total suckling time, and the mean suckling time during the first 90 min after fasting were not different in males and females and were not altered differently by placental restriction or birth weight (data not shown).
Total suckling time, in absolute terms (r = −0.67, n = 7, P = 0.05) and adjusted to current weight (r = −0.71, n = 7, P = 0.025), in the third 15-min period after the cessation of fasting correlated negatively with birth weight (Fig. 5). Total suckling time, in absolute terms (r = −0.78, n = 7, P = 0.019) and adjusted to current weight (r = −0.80, n = 7, P = 0.015), in the third 15-min period after the cessation of fasting correlated negatively with abdominal circumference at birth (Fig. 5).
Feeding activity in terms of mean suckling time (throughout the 90 min after fasting) correlated negatively with CRL at birth (r = −0.76, n = 7, P = 0.023). Mean suckling time (throughout the 90 min after fasting) adjusted for current weight also correlated negatively with CRL (r = −0.93, n = 7, P = 0.001), tibia length (r = −0.79, n = 7, P = 0.016), abdominal circumference (r = −0.82, n = 7, P = 0.011; Fig. 6), and metatarsal length (r = −0.69, n = 7, P = 0.042) at birth.
Feeding activity, growth, and adiposity.
Total suckling time correlated positively with CFGR in terms of weight (r = 0.70, n = 7, P = 0.041; day 15). Similarly, total suckling time (during the third 15-min period after fasting) correlated positively with CFGR in terms of CRL (r = 0.80, n = 7, P = 0.016), metatarsal length (r = 0.69, n = 7, P = 0.045), and skull length (r = 0.70, n = 7, P = 0.039; day 15). The weight gain from 30 to 150 min after the cessation of fasting correlated positively with CFGR for weight (r = 0.74, n = 7, P = 0.028; day 15).
The weight gain from 30 to 150 min after the cessation of fasting correlated positively with left retroperitoneal fat weight (r = 0.73, n = 7, P = 0.031), right retroperitoneal fat weight (r = 0.77, n = 7, P = 0.023), total retroperitoneal fat weight (r = 0.75, n = 7, P = 0.026), and total retroperitoneal fat weight as percentage of live weight (r = 0.76, n = 7, P = 0.023) in all animals.
In this study, experimental restriction of placental implantation and growth in the sheep reduced placental weight and size of offspring at birth and induced catch-up growth in terms of weight and length in the first few weeks of life. Furthermore, placental restriction reduced skeletal muscle mass and increased central adiposity in absolute and relative terms in lambs at 1 mo of age. Therefore, experimental placental insufficiency in the sheep induced accelerated growth in terms of weight and length and early-onset central obesity after birth, consistent with observed associations with IUGR in humans (8, 39). In addition, placental restriction altered feeding activity in the young lamb at 2 wk of age, appearing to induce a more sustained suckling effort in terms of suckling time after fasting. Furthermore, this increased feeding activity was predictive of catch-up growth and increased central adiposity in the young lamb. This suggests that placental substrate deprivation before birth, evident as small size at birth, alters early postnatal growth and body composition, in part, by programming increased feeding activity.
Here, in a larger cohort than that used in our previous study (17), we have demonstrated that restricted implantation and growth of the placenta induce early postnatal catch-up growth, which is complete in terms of most parameters by 45 days of age in the young lamb. Increased NFGR and CFGR indicate higher growth rates relative to initial size and, hence, reflect a greater partitioning of nutrients toward growth. Fractional growth rates were increased consistently across measures of size by placental restriction. Although AGR were generally not increased after placental restriction, placentally restricted lambs had caught up in terms of most size parameters by 35 days of postnatal age. An increase in neonatal growth rate in terms of weight, height, and head size occurs in most human infants after IUGR in the first weeks and months of life (3, 31, 33), but, similar to our present observations in the sheep, catch-up growth in terms of head size in humans is incomplete at 6 mo of age (37). Some catch-up growth in terms of length of long bones also occurred in the young lamb after placental restriction, although the consequences in terms of bone density and composition in the young lamb were not determined. Spontaneous low birth weight in lambs due to multiple births reduces ash content in ground body tissue, but the effect on bone growth was not directly determined (27). Nevertheless, together with our observations, this suggests that bone may be more limited than soft tissue in its capacity to catch up after birth, even when nutrient availability is unlimited (27).
In a previous study, placental restriction reduced absolute perirenal adipose tissue weight and tended to reduce perirenal adipose tissue weight adjusted for body weight in the near-term fetal sheep (54), suggesting that the increased central fat accumulation we observed in such offspring at 6 wk of age is of postnatal origin. In the present study as well as in a smaller subset of these animals (17), placental restriction increased central fat mass (visceral fat) in absolute and relative terms in the young lamb, and this was predicted by increased growth rate in terms of weight. These observations are consistent with human studies in which IUGR infants undergo accelerated growth for their size during the first few months of life (4, 31, 33, 34, 36, 37, 48) and have an increased risk of impairment of later function and health, including obesity (8, 45). Epidemiological studies have also shown that early postnatal catch-up growth is an independent risk factor for obesity in children and adults (2, 39, 46). Such early onset and long-term persistence of increased adiposity after placental restriction and IUGR may contribute to their increased risk of developing obesity-related disorders, including cardiovascular disease, later in life (6). Importantly, the present study suggests that intervention to prevent or ameliorate this early onset of central obesity after IUGR may be possible early in postnatal life.
In the present study, reduced size at birth in terms of a range of parameters was predictive of reduced individual and summed skeletal muscle mass in absolute terms and relative to body weight in the young lamb. This relation was evident for muscles that are composed predominantly of type I (tibialis, semitendinosus, soleus, and biceps femoris) and type II (biceps, gastrocnemius, extensor digitorum longus, and vastus lateralis) fibers in sheep (22, 32). This suggests that lean tissue growth is restrained or limited after IUGR and, especially, relative to the growth of adipose tissue. Muscle of spontaneously low-birth-weight lambs has reduced DNA content and a higher protein-to-DNA ratio than muscle of high-birth-weight lambs, suggesting reduced myonuclei number as a possible explanation for a limited capacity for postnatal growth of skeletal muscle after IUGR (28). This may exacerbate the relatively increased adiposity of the IUGR lambs undergoing catch-up growth (28, 51).
The mechanistic basis of catch-up growth and increased visceral adiposity after IUGR are poorly understood, but programming of altered activity of neuroendocrine and endocrine axes, which influence growth, appetite, satiety, and metabolic efficiency, is implicated in the phenomenon (8, 15, 17, 21, 25, 53, 59). Here we show that reduced size at birth is associated with acutely altered and overall increased feeding activity in terms of total and mean suckling times after the cessation of short-term fasting. There appeared to be regular peaks in feeding activity after fasting in young lambs, with a more sustained suckling effort in terms of suckling time associated with placental restriction or small size at birth. Hyperphagia occurs in the adult rat (57) and guinea pig (35) after spontaneous or experimental restriction of fetal substrate supply. Our observations indicate that prenatal restriction also induces hyperphagia soon after birth, during the period of maximal catch-up growth. Similarly, Greenwood et al. (27) reported that feed intake was higher in low- than in high-birth-weight suckling lambs and that low birth weight and ad libitum feeding each increased neonatal growth and adiposity in sheep. In the present study, feeding activity predicted NFGR and visceral fat mass, suggesting that increased feeding activity may contribute to catch-up growth and development of central obesity in the placentally restricted lamb or the lamb that was small at birth. Together, these studies suggest that altered appetite and/or satiety may contribute to accelerated growth and obesity after restriction of prenatal growth. The milk composition of the ewe and quantity ingested by the lamb during feeding were not measured in the present study. Whether this increased feeding activity in the prenatally restricted young lamb results in increased feed intake and nutrient absorption therefore remains to be determined, as do the mechanisms underlying changes in feeding behavior.
In summary, placental restriction reduces size at birth, which is associated with increased feeding activity, and induces neonatal catch-up growth in the young lamb. Although catch-up growth following placental restriction normalizes size in terms of most parameters after 1 mo of age, skeletal muscle mass is reduced and central adipose tissue is increased in these young lambs. Catch-up growth and increased central adiposity in the young lamb that was small at birth are predicted by, and may be due in part to, increased feeding activity. If these adverse alterations in body composition persist, as they do in other species, after IUGR, they may contribute to adverse metabolic and cardiovascular outcomes in later life. Importantly, the early postnatal onset of central obesity after IUGR suggests that intervention at this stage of development to prevent or ameliorate this may be possible.
This project was funded by a grant from the National Health and Medical Research Council of Australia Program. K. L. Gatford was supported by a Peter Doherty Postdoctoral Fellowship from the National Health and Medical Research Council of Australia.
We thank Frank Carbone, Dr. Tim Butler, Anne Jurisevic, and Dr. Mike Adams for surgical assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society