PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Behav Brain Res. Author manuscript; available in PMC Jun 26, 2011.
Published in final edited form as:
PMCID: PMC2866501
NIHMSID: NIHMS183341
One Week of Exposure to Intermittent Hypoxia Impairs Attentional Set-shifting in Rats
John G. McCoy,1,2 James T. McKenna,1 Nina P. Connolly,1,3 Devon L. Poeta,2 Liming Ling,4 Robert W. McCarley,1 and Robert E. Strecker1
*1VA Boston Healthcare System and Harvard Medical School, Laboratory of Neuroscience, Research 151-C, 940 Belmont Street, Brockton, MA 02301, USA
2Stonehill College, Department of Psychology, Easton, MA 02357, USA
3Wheaton College, Department of Psychology, Norton, MA 02766, USA
4Division of Sleep Medicine, Brigham and Womens Hospital, Harvard Medical School, Boston, Massachusetts 02115, USA
*1Institution where work was performed.
Abstract
Intermittent hypoxia (IH), a characteristic of sleep apnea, was modeled in Fischer Brown Norway rats (10 h/day for 7 days) followed by cognitive testing in an attentional set-shifting task. The ability to shift attention from one sensory modality (e.g., odor) to another (e.g., digging medium) was impaired, a finding that could not be attributed to deficits in attention, discrimination, learning, or motor performance. Instead, the deficit is likely to reflect impaired allocation of attentional resources of the working memory system.
Keywords: attention, hypoxia, sleep apnea syndromes, discrimination learning, animal models
Intermittent hypoxia (IH) and sleep fragmentation are primary characteristics of the sleep apnea syndromes. Adult patients with sleep apnea typically exhibit excessive daytime sleepiness, mood disturbances, and impaired cognition (Roehrs et al., 1995). Evidence suggests that both IH (Row, 2007) and sleep fragmentation (Thomas et al., 2005) may contribute significantly to the observed cognitive deficits. Impaired memory has been observed in humans diagnosed with sleep apnea (Bedard et al., 1991), as well as in laboratory rats exposed to IH to model sleep apnea (Gozal, Daniel, and Dohanich, 2001). While spatial memory deficits in rodents have been repeatedly observed following IH (See Row, 2007 for a review), fewer attempts have been made to model the deficits in executive functioning commonly seen in apneic patients.
In the present study, laboratory rats were tested in an attentional set-shifting task following one week (10 h/day for 7 days) of exposure to IH. The task used here was developed by Birrell and Brown (2000) to provide a rodent-appropriate version of the human intradimensional/extradimensional (ID/ED) shifting task (Owen et al., 1991), a test similar in method and purpose to the more familiar Wisconsin Card Sorting Test (WCST) for humans. The rationale for employing the attentional set-shifting task is based on evidence of dysregulation of inputs to the prefrontal cortex (PFC) in humans diagnosed with sleep apnea (Thomas et al., 2005; Yaouhi et al., 2009; Macey et al., 2008).
The PFC is well known to play a central role in “executive functioning,” a term which refers here to the parceling out of attentional resources in response to changing environmental demands by components of the working memory system, which holds information on-line for immediate use (Baddeley, 1992). Studies employing positron emission tomography (PET) have shown that extradimensional shifting activates the dorsolateral PFC in humans (Rogers et al., 2000). Furthermore, humans who have sustained damage to the PFC exhibit impaired performance on problem solving tasks that require the subject to shift attention from one rule to another (Owen et al., 1991). In rats, lesion studies have demonstrated a key role for the medial PFC in attentional set-shifting (Birrell and Brown, 2000). The present study evaluated the performance of rats on the ID/ED attentional set-shifting task following exposure to one week of IH (10 h/day for 7 consecutive days).
Fourteen adult male Fischer Brown Norway F1 rats (252–274 g; Harlan Laboratories, Ltd.) were first habituated to the custom designed cages (l × w × h = 35.5 cm × 22.8 cm × 20.3 cm) with room air being infused into the chamber through two alternating air sources in order to mimic the experimental conditions of alternating flow rates of N2 and normal air described below. The habituation period to room air lasted for 48 h prior to the IH exposure. Rats were exposed to the IH protocol for 10h/day (from 2pm to midnight) for 7 consecutive days (lights on at 8 AM, off at 8 PM). With this IH exposure schedule, the morning hours were available to habituate the rats to the IDED task apparatus (days 1 & 2) and to shape the rats to find food pellets buried in terra cotta pots (3 to 5 shaping sessions occurred on days 1 to 6; see McCoy et al. 2007 for details). On the morning of day 7, the rats experienced exemplar training to acquaint the animals with the possible reward-response contingencies of the task (McCoy et al. 2007), and then were returned to the IH cages for the final 10 hours of IH exposure. The formal IDED task occurred on the morning of day 8.
The IH regimen is a modified version of the protocol originally developed (Gozal et al., 2001) and previously described by us (Ward et al., 2009). The custom designed cages restricted airflow and were designed to allow the cage O2 levels to be systematically varied. Infused gas cycled inside the cage from room air (21% O2 at 11 L/min for 60s) to nitrogen rich air (6% O2 at 8L/min for 60 s). Cage O2 levels were measured via an oximeter. Thus, rats were exposed to environmental oxygen on a schedule that produced hypoxia with a similar frequency observed in typical human with sleep apnea (30 hypoxic episodes/h); in rats this IH regimen does not significantly alter sleep after the first 24h of exposure (Gozal et al., 2001). The resultant hypoxemia that mimicked the blood O2 de-saturation typical of sleep apnea (70 to 75% O2 de-saturation relative to normal blood O2 using methods previously described by McGuire et al. (2002). This exposure produced at least 20 s of inspired air levels below 10%, followed by 60 s of normal air, which produced at least 20 s of inspired air levels above 18% O2. Rats had free access to water in the cages. Air control rats lived in a similar cage with identical flow rates of room air infusion (11 L/min for 60 s and 8L/min for 60 s).
The attentional set-shifting task and food restriction procedure for rats has been described in detail previously (McCoy et al., 2007; Birrell and Brown, 2000). Rats were ordered weighing 220 g, allowed food al libitum until they reached at least 250 g. Each rat had an initial body weight between 250 and 275 g at the start of the food restriction protocol. Rats were then given 12 g of dry food pellets per day until they reached 90% of their initial body weight (typically 9 to 14 days of food restriction). Rats were weighed daily and rats that went below their 90% threshold were given additional daily food. Food restriction continued during the 7 days of IH and rats were fed 12 g food per day prior to the daily IH exposure (~1:00p.m.). Rats obtained additional food in the habituation, shaping, and IDED experimental procedures. Body weight relative to each rat’s initial body weight on the last day of the study ranged from 89.5 to 93% (mean = 90.6%). Rats appeared healthy and groomed normally throughout the course of this experiment suggesting that stress or other non-specific effects of combining IH exposure and moderate food restriction were unlikely to produce the specific behavioral impairments observed.
Rats were trained to discriminate between two terra cotta pots to obtain a small food reward (~0.05–0.15 g bit of rat chow) for each correct response. The pots were placed within the goal box portion of the test apparatus, a large, clear plastic box (16 cm tall, 90 cm long, and 44 cm wide). An opaque, removable divider separated the goal box from the remainder of the apparatus, which functioned as the starting point for each rat. Testing in the set-shift apparatus was conducted during the lights on phase. The task involved 2 pairs of stimuli to be discriminated. Each discrimination was represented by the pair of pots (Table 1). One pot was deemed the “correct” choice, based on either digging medium (e.g., different color paper, beads, etc. that filled the pots) or odor (i.e., scented oils applied to the rim of the pots). See McCoy et al. (2007) for a list of the specific digging media and odors that were used. To prevent rats from using olfactory cues to signal the correct pot, a small amount of powdered food was placed in the incorrect pot.
Table 1
Table 1
Typical example of trial pairings for rat during attentional set-shifting task
Following habituation and shaping (see McCoy et al., 2007 for details), animals were tested in the following series of discrimination tests: simple discrimination, compound discrimination, reversal 1, intradimensional shift, reversal 2, extra-dimensional shift, reversal 3. For each discrimination test, animals were required to reach a criterion of 6 consecutive correct responses (referred to as “trials-to-criterion”) before moving on to the next discrimination. A correct response was determined by direct observation of digging in the pot containing the reinforcement. Prior digging in the unrewarded pot constituted an incorrect response and ended the trial. In the simple discrimination, one of the two dimensions (e.g., odor in Table 1) were held constant while the other was varied (i.e., brown versus white paper) with one of the two (e.g., brown) being reinforced. In the complex discrimination, both dimensions were varied, but only one dimension was relevant and reinforced. Thus, rats learned to discriminate between stimuli within the relevant dimension (e.g., digging medium) while disregarding the other dimension (e.g., odor). For reversals, the previously unattended stimulus was now reinforced (e.g., white paper). For the intradimensional shift, animals were required to apply the previously learned rule regarding which stimulus dimension predicts reward to two novel stimuli (see Table 1). For the extradimensional shift, animals were required to shift their attention away from the previously reinforced stimulus dimension (i.e., digging medium) to the previously irrelevant dimension (i.e., a specific odor is now reinforced). The literature indicates that the extradimensional shift is the component that is most susceptible to disruption, with deficits on the ED shift observed following numerous manipulations, including brain lesions, sleep manipulations, pharmacological treatments, etc. (see Danet et al. 2008 for review).
Behavioural performance on the ID/ED task was analyzed using one-way analyses of variance (ANOVA) for independent groups (i.e., IH-exposed and room air controls). A separate one-way ANOVA was conducted for each of the 7 discrimination tests of the attentional set-shifting task. All analyses were performed using SPSS (Version 12) for Windows.
Following 7 consecutive days (10 h/day) of IH, a selective impairment was found in the extradimensional shift component of the ID/ED attentional set-shifting task (F(1,12)=19.15, P=0.001). IH-exposed animals required an average of 17.9 (± 1.3) trials to reach criterion, while room air controls reached criterion in only 10.3 (± 0.9) trials. This equates to nearly a 58% increase in trials required to reach criterion when making an extra-dimensional shift (see Figure 1). There were no impairments on any of the other phases of the ID/ED task.
Figure 1
Figure 1
Exposure of rats to intermittent hypoxia 10 h per day for 7 consecutive days selectively impairs performance on the extradimensional shift (ES) discrimination of the ID/ED attentional set-shifting task. Data are expressed as the mean (±SEM) trials-to-criterion (more ...)
The intermittent hypoxia (IH) paradigm employed in the present study was adapted from the protocol used by Gozal and associates (2001) to model the episodic occurrence of hypoxia which characterizes the sleep apnea syndromes. We have previously reported that 3 d of IH (10 h/day) impaired spatial learning of rats in the Morris water maze (Ward et al., 2009). We now report additional impairments in attentional set-shifting following a longer duration of IH (7 days for 10h/day). Since significant differences were not found on the intradimensional shift component, the observed impairment cannot be interpreted as a general deficit in the ability to solve new discrimination problems. A general deficit in attention, cognition or motor performance is also unlikely, as general deficits would have affected multiple components of the ID/ED task. The lack of an effect on reversal learning rules out the possibility that animals exposed to IH might merely be perseverating. Rather, one week of exposure to IH specifically impaired the ability of rats to shift attentional set from one sensory modality (e.g., odor) to another (e.g., digging medium). This selective finding suggests that IH impairs the allocation of attentional resources of the working memory system that are associated with the concept of executive function and the pre-frontal cortex.
It is of interest that both 24 h of sleep fragmentation (McCoy et al., 2007) and the 7 days of IH exposure used herein produce very similar and specific deficits on the ED shift component of the ID/ED task. The severity of the impairment was similar in both groups of experimental animals with rats exposed acute (i.e., 24 h) sleep fragmentation requiring nearly as many trials to reach criterion (15.4 trials) as the rats exposed to one week of IH in the present study (17.9 trials). We hypothesize that this is because both acute sleep disruption and sub-chronic IH exposure interfere with the function of forebrain structures that are involved in the allocation of the working memory resources needed to perform the ED shift in the ID/ED attentional set shifting task.
While the only significant IH-induced impairment occurred on the extradimensional shift in the present study, an overall trend toward diminished performance (i.e., more trials to reach criterion in IH-exposed animals) is noticeable across almost all components of the ID/ED task. Our preferred explanation for this trend is that IH-exposed animals were noticeably hyperactive during the test session, making it more difficult to focus attention on the task at hand. For example, IH animals often attempted to jump out of the test apparatus, whereas control animals exhibited no such behaviour. Our observations are consistent with reports of enhanced locomotor activity (Decker et al., 2003,2005;Row et al., 2002) and executive dysfunction (Decker et al., 2003) in rats following IH exposure. In these studies, rats were exposed to IH during critical early developmental periods in order to model hyperactivity and attentional deficits noted in children with obstructive sleep apnea, or other forms of sleep disordered breathing (Hallbower and Mahone, 2006). We now report executive dysfunction and hyperactivity (although this was not quantified herein) in rats exposed to IH at a mature age (approximately 100–150 days).
The capacity to disregard the originally relevant dimension, and attend to a new (formerly irrelevant) dimension is considered an executive function of the nervous system. Converging evidence strongly suggests that the PFC plays an important role in executive processes not only in human (Owen et al., 1991) and nonhuman (Dias et al., 1997) primates, but also in laboratory rats; thus, lesions to the medial PFC in rats have been shown to impair extradimensional shifting on the ID/ED task (Birrell and Brown, 2000). Changes in frontal lobe perfusion (Alchanatis et al., 2004) as well as deficits in executive functioning (Fulda and Schultz, 2003) have been documented in humans diagnosed with obstructive sleep apnea including attentional set-shifting (Wong et al., 2006) and working memory (Thomas et al., 2005). These deficits are associated with a disproportionate impairment of function in the dorsolateral prefrontal cortex (Thomas et al., 2005). Specifically, a meta-analysis revealed the most profound cognitive impairments in sleep-disordered patients to be those that involve functions of the PFC. These functions include sustained attention and working memory tasks that require mental flexibility, such as the WCST (Fulda and Schultz, 2003). Thus, dysregulation of the PFC in humans diagnosed with sleep apnea is thought to underlie, at least in part, the impaired cognitive and executive functioning observed in these patients. However, contrary to these reports of deficits, a recent report on healthy adult humans exposed experimentally to four weeks of IH revealed no significant effects on alertness, vigilance, or working memory (Weiss et al., 2009).
The mechanism(s) underlying the IH-induced attentional deficit is presently unknown. Chronic intermittent hypoxia has been shown to induce oxidative stress and neuronal apoptosis (Xu et al. 2004). Certain brain regions, such as the hippocampus and cerebellum, are known to be more sensitive to (the after-effects of) hypoxia than others (Cervos-Navarro and Diemer, 1991). Task-related activations might also render additional brain regions sensitive to such effects. Another possibility involves basal forebrain cholinergic projections to the PFC. Fourteen days of IH in rats resulted in working memory impairments, along with reductions in choline acetyltransferase in the medial septum and the substantia inominata, as well as increases in nicotinic cholinergic binding sites in the PFC (Row et al., 2007). Whether the attentional effects of IH might also be mediated by cholinergic mechanisms remains an open question.
There is some possibility that the effects observed herein are the result of disruptions in sleep, which have been documented following chronic IH in mice (Veasey et al., 2004; Plotsky et al., 2006). Although sleep disruption has also been demonstrated in rats during the first day of IH, sleep has been found to be normalized after the first day of exposure in rats (Gozal et al., 2001, and our unpublished observations). More importantly, in the present study, rats were only exposed to 10h/day of IH, followed by 12 to 14 hours of normoxia allowing sleep without the potential disruptions caused by the IH procedure (air flow noise; changes in oxygen level); this daily period of normoxia always immediately preceded any training or testing in the ID/ED task. Therefore, it is unlikely that the behavioral impairments observed herein are due to sleep disruption. However, sleep recordings, suggested as a future direction, are needed to be 100% certain that the IH regimen used herein does not disrupt sleep.
The deficit reported in rats in the present study should also be viewed with some caution as deficits in the ED shift component of the ID/ED task have been reported following a variety of manipulations (e.g., sleep fragmentation, brain lesions, pharmacological manipulations). Thus, a lack of behavioral specificity for each component of the ID/ED task is a potential concern. However, selective deficits on specific IDED task components have been reported by others (McAlonan and Brown, 2003; including an elegant series of experiments by Morilak and colleagues demonstrating bidirectional control of different elements of the ID/ED task by using medial prefrontal cortex microinjections of noradrenergic agonists and antagonists (Lapiz et al., 2006). To address this issue the findings of the present study need to be investigated using neurobiological methods, such as lesions or injections of drugs in specific brain regions.
In summary, exposure to IH (10 h/day for 7 consecutive days) impaired attentional set-shifting in young adult male rats. Additionally, a trend towards diminished performance was noted on the ID/ED task, presumably due to hyperactivity observed in IH animals only. The IH protocol used here produced behavioural effects similar to those which characterize sleep apnea in humans. Finally, the ID/ED attentional set-shifting task provides a useful method to assess central executive function, and its underlying neural mechanisms, in rodents.
Acknowledgements
This research was supported by the Department of Veteran’s Affairs (RES and RWM), NIH HL060292 (RWM and RES), NIH T32 HL07901 (JTM) and NIH F32 MH070156 (JTM).
Footnotes
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
  • Alchanatis M, Del giorgis N, Zias N, Amfilochiou A, Gotsis E, Karakatsani A, Papadimitriou A. Frontal brain lobe impairment in obstructive sleep apnea: a proton MR spectroscopy study. Eur Respir J. 2004;24:980–986. [PubMed]
  • Baddeley A. Working memory. Science. 1992;255:556–559. [PubMed]
  • Bedard MA, Montplaisir J, Richer F, Rouleau I, Malo J. Obstructive sleep apnea syndrome: pathogenesis of neuropsychological deficits. J Clin Exp Neuropsychol. 1991;13:950–964. [PubMed]
  • Birrell JM, Brown VJ. Medial frontal cortex mediates perceptual attentional set shifting in the rat. J Neurosci. 2000;20:4320–4324. [PubMed]
  • Cervos-Navarro J, Diemer NH. Selective vulnerability in brain hypoxia. Crit Rev Neurobiol. 1991;6:149–182. [PubMed]
  • Danet M, Lapiz-Bluhm S, Bondi CO, Doyen J, Rodriguez G, Bedard-Arana T, Morilak DA. Behavioural assays to model cognitive and affective dimensions of depression and anxiety in rats. J Neuroendocrinol. 2008;20:1115–1137. [PMC free article] [PubMed]
  • Decker MJ, Hue GE, Candle WM, Miller GW, Keating GL, Rye DB. Episodic neonatal hypoxia evokes executive dysfunction and regionally specific alterations in markers of dopamine signaling. Neuroscience. 2003;117:417–425. [PubMed]
  • Decker MJ, Jones RA, Solomon JG, Keating GL, Rye DB. Reduced extracellular dopamine and increased responsiveness to novelty: neurochemical and behavioral sequelae of intermittent hypoxia. Sleep. 2005;28:169–176. [PubMed]
  • Fulda S, Schultz H. Cognitive dysfunction in sleep-related breathing disorders: a meta-analysis. Sleep Res On-line. 2003;5:19–51.
  • Gozal D, Daniel JM, Dohanich GP. Behavioral and anatomical correlates of chronic episodic hypoxia during sleep in the rat. J Neurosci. 2001;21:2442–2450. [PubMed]
  • Halbower AC, Mahone EM. Neuropsychological morbidity linked to childhood sleep disordered breathing. Sleep Med Rev. 2006;10:97–107. [PubMed]
  • Lapiz MD, Morilak DA. Noradrenergic modulation of cognitive function in rat medial prefrontal cortex as measured by attentional set-shifting capability. Neuroscience. 2006;137:1039–1049. [PubMed]
  • Macey PM, Kumar R, Woo MA, Valladares EM, Yan-Go Fl, Harper RM. Brain structural changes in obstructive sleep apnea. Sleep Res. 2008;31:967–977. [PubMed]
  • McAlonan K, Brown VJ. Orbital prefrontal cortex mediates reversal learning and not attentional set-shifting in the rat. Behav Brain Res. 2003;146:97–103. [PubMed]
  • McCoy JG, Tartar JL, Bebis AC, Ward CP, McKenna JT, Baxter MG, McGaughy J, McCarley RW, Strecker RE. Experimental sleep fragmentation impairs attentional set-shifting in rats. Sleep. 2007;30:52–60. [PubMed]
  • McGuire M, Zhang Y, White DP, Ling L. Effect of hypoxic episode number and severity on ventilatory long-term facilitation in awake rats. J Appl Physiol. 2002;93:2155–2161. [PubMed]
  • Owen AM, Roberts AC, Polkey CE, Sahakian IH, Robbins TW. Extra-dimensional versus intra-dimensional set shifting performance following frontal lobe excisions, temporal lobe excisions or amygdalo-hippocampectomy in man. Neuropsychologia. 1991;29:993–1006. [PubMed]
  • Polotsky VL, Rubin AE, Balbir A, Dean T, Smith PL, Schwartz AR, O'Donnell CP. Intermittent hypoxia causes REM sleep deficits and decreases EEG delta power in NREM sleep in the C57 BL/6T mouse. Sleep Med. 2006;7:7–16. [PubMed]
  • Roehrs T, Merrion M, Pedrosi B, Stepanski E, Zorick F, Roth T. Neuropsychological function in obstructive sleep apnea syndrome (OSAS) compared to chronic obstructive pulmonary disease (COPD) Sleep. 1995;18:382–388. [PubMed]
  • Rogers RD, Andrews TC, Grasby PM, Brooks DJ, Robbins TW. Contrasting cortical and subcortical activations produced by attentional-set shifting and reversal learning in humans. J Cogn Neurosci. 2000;12:142–162. [PubMed]
  • Row BW. Intermittent hypoxia and cognitive function: implications from chronic animal models. Adv Exp Med Biol. 2007;618:51–67. [PubMed]
  • Row BW, Kheirandish L, Cheng Y, Rowell PP, Gozal D. Impaired spatial working memory and altered choline acetyltransferase (CHAT) immunoreactivity and nicotinic receptor binding in rats exposed to intermittent hypoxia during sleep. Behav Brain Res. 2007;177:308–314. [PMC free article] [PubMed]
  • Row BW, Kheirandish L, Neville JJ, Gozal D. Impaired spatial learning and hyperactivity in developing rats exposed to intermittent hypoxia. Pediatr Res. 2002;52:449–453. [PubMed]
  • Thomas RJ, Rosen BR, Stern CE, Weiss JW, Kwong KK. Functional imaging of working memory in obstructive sleep-disordered breathing. J Appl Physiol. 2005;98:2226–2234. [PubMed]
  • Veasey SC, Davis CW, Fenik P, Zhan G, Hsu YJ, Practico D, Gow A. Long-term intermittent hypoxia in mice: protracted hypersomnalence with oxidative injury to sleep-wake brain regions. Sleep. 2004;27:194–201. [PubMed]
  • Ward CP, McCoy JG, McKenna JT, Connolly NP, McCarley RW, Strecker RE. Spatial learning and memory deficits following exposure to 24 hours of sleep fragmentation or intermittent hypoxia in a rat model of obstructive sleep apnea. Brain Research. 2009;1294:128–137. [PMC free article] [PubMed]
  • Weiss MD, Tamisier R, Boucher J, Lynch M, Gilmartin G, Weiss JW, Thomas RJ. A pilot study of sleep, cognition and respiration under 4 weeks of intermittent nocturnal hypoxia in adult humans. Sleep Med. 2009;10:739–745. [PubMed]
  • Wong KK, Grunstein RR, Bartlett DJ, Gordon E. Brain function in obstructive sleep apnea: results from the Brain Resource International Database. J Integr Neurosci. 2006;5:111–121. [PubMed]
  • Xu W, Chi L, Row BW, Xu R, Ke Y, Xu B, Luo C, Kheirandish L, Gozal D, Liu R. Increased oxidative stress is associated with chronic intermittent hypoxia-mediated brain cortical neuronal cell apoptosis in a mouse model of sleep apnea. Neuroscience. 2004;126:313–323. [PubMed]
  • Yaouhi K, Bertran F, Clochon P, Mezenge F, Denise P, Foret J, Eustache F, Desqranges B. A combined neuropsychological and brain imaging study of obstructive sleep apnea. J Sleep Res. 2009;18:36–48. [PubMed]