Previously presented in the form of a poster at the 25th Symposium of the Veterinary Comparative Respiratory Society, Lafayette, IN.
Corresponding author: Undine Christmann, DVM, MS, PhD, DACVIM, Department of Large Animal Clinical Sciences, Virginia-Maryland Regional College of Veterinary Medicine (VMRCVM), Duck Pond Drive, Phase II, Blacksburg, VA 24061-0442; e-mail: email@example.com.
Background: Surfactant alterations are described in horses after exercise, anesthesia, and prolonged transport, in horses with recurrent airway obstruction, and in neonatal foals. The effect of horse age or bronchoalveolar lavage fluid (BALF) sample characteristics on surfactant is unknown.
Objectives: To evaluate surfactant phospholipid composition and function in healthy horses, and to investigate the influence of age and BALF sample characteristics on surfactant.
Animals: Seventeen healthy horses 6–25 years of age maintained on pasture year-round.
Methods: BALF was collected by standard procedures and was assessed for recovery volume, nucleated cell count (NCC), and cytology. Cell-free BALF was separated into crude surfactant pellet (CSP) and surfactant supernatant (Supe) by ultracentrifugation. Phospholipid and protein content were determined from both fractions. CSP phospholipid composition was analyzed by high-performance liquid chromatography with an evaporative light scatter detector. Surface tension of CSP was evaluated with a pulsating bubble surfactometer. Regression analysis was used to evaluate associations between age, BALF sample characteristics, and surfactant variables.
Results: Results and conclusions were derived from 15 horses. Increasing age was associated with decreased phospholipid content in CSP but not Supe. Age did not affect protein content of CSP or Supe, or surfactant phospholipid composition or function. Age-related surfactant changes were unaffected by BALF recovery percentage, NCC, and cytological profile.
Conclusions and Clinical Importance: Older horses have decreased surfactant phospholipid content, which might be because of age-related pulmonary changes. Surfactant composition is unaffected by BALF sample characteristics at a BALF recovery percentage of at least 50%.
total phospholipid content in cell-free BALF in μg/mL
protein content in CSP in μg/mL
protein content in Supe in μg/mL
protein content in cell-free BALF in μg/mL
ratio between protein content and phospholipid content in cell-free BALF
Pulmonary surfactant lowers surface tension (biophysical function) at the level of the alveoli and airways and modulates pulmonary immune responses (immunological function).1–3 The majority of surfactant (>80%) is composed of phospholipids that are largely responsible for its biophysical function. Surfactant contains approximately 12% of the following surfactant proteins: SP-A, SP-B, SP-C, and SP-D. SP-A and SP-D are part of the innate immune defenses of the lung,4 whereas SP-B and SP-C contribute to biophysical surfactant function.5 Surfactant also contains a small percentage of neutral lipids (eg, cholesterol, triglycerides), the role of which is not fully characterized. Surfactant is synthesized, stored, and secreted by type II alveolar cells.6 The majority of surfactant undergoes extensive recycling; a small part of surfactant is moved toward the airways, and another fraction is degraded by alveolar macrophages.7,8
Surfactant alterations play a role in alveolar and airway disease. In the alveoli, low surfactant surface tension is essential to prevent alveolar collapse.1,3 Surfactant deficiency in premature animals with immature lungs leads to neonatal respiratory distress syndrome.9 Surfactant dysfunction, evident as high surface tension, contributes to the pathophysiology of acute respiratory distress syndrome.10,11 In the airways, surfactant helps maintain patency of small airways,12 improves mucociliary clearance,13 and decreases bronchoconstriction in response to inhaled allergens.14,15 Alterations in surfactant composition and function have been described in a variety of airway diseases.1,16 For example, in human medicine, surfactant alterations occur in patients with allergic asthma,15,17,18 cystic fibrosis,19 chronic bronchitis,20 and viral respiratory diseases.1,3 Surfactant alterations have been reported in horses after transport,21,22 exercise,23 and general anesthesia,24 in horses with recurrent airway obstruction,25,26 and in neonatal foals.27
Surfactant changes associated with pulmonary maturation are well documented in neonates from a number of animal species,28–30 including the horse.27 However, only a few reports exist on age-related changes occurring after the neonatal period.31–35 In addition, the effect of bronchoalveolar lavage fluid (BALF) sample characteristics (eg, BALF recovery, cell count, and cytological profile) has not been investigated in the horse. The purpose of our study was (1) to measure surfactant composition and function in healthy horses, and (2) to investigate the influence of age and BALF sample characteristics on surfactant variables.
Material and Methods
Seventeen horses from the Virginia-Maryland Regional College of Veterinary Medicine (VMRCVM) teaching herd were selected for this study. Horse breeds included 8 Thoroughbreds, 4 American Quarter Horses, 1 Arab, 1 Spanish Barb, and 3 mixed breed horses. Fourteen of the horses were mares and 3 were geldings. Ages ranged from 6 to 25 years of age (mean ± standard deviation [SD], 13 ± 6 years) and body weights ranged from 427 to 625 kg (mean ± SD, 538 ± 54 kg). Horses were housed on adjacent pastures (4–5 horses per pasture) year-long and did not have a history or clinical signs of respiratory disease for at least 1 year. All horses were routinely dewormed and vaccinated. During winter months, their diet was supplemented with hay. Criteria for enrollment in the study were normal findings on clinical examination, lung auscultation with a rebreathing bag, and CBC as well as absence of airway obstruction after exposure to barn environment. Samples were collected in June and July 2005. On the day before sampling, horses were brought into the research barn where they were bedded on shavings and fed free-choice hay off the ground. Each horse had free access to a small individual outdoor pen. All procedures were approved by the Institutional Animal Care and Use Committee at the VMRCVM.
BALF Collection and Analysis
Horses were held in stocks and sedated by administration of detomidine hydrochloridea (0.01 mg/kg IV) and butorphanol tartrateb (0.01 mg/kg IV). A BALF collection tubec was passed through the nasal passages, and into the trachea and lower airways. Coughing was inhibited by spraying the airways with a 0.2% lidocaine solution as the BALF collection tube was passed. Once the tube was wedged, BALF was collected by instilling 300 mL of prewarmed sterile saline solution and by reaspirating BALF with a syringe. Samples from each horse were pooled in a sterile specimen cup, mixed, placed on ice, and processed within 30 minutes after collection. Part of the sample was submitted for cell count and cytologic examination. The rest of the sample was centrifuged for 10 minutes at 400 ×g and 4°C, the cell-free supernatant was stored at − 80°C until analysis.
BALF cell counts were determined with an automated cell counter.d For BALF cytology, 300 μL of each sample were concentrated by cytospin. Slides were stained by use of a quick Wright's stain. A 400-cell leukocyte differential count was performed on all slides.
Surfactant Isolation and Analysis
Cell-free BALF supernatant was centrifuged at 45,000 ×g for 1 hour to allow separation of crude surfactant pellets (CSP) and surfactant supernatant (Supe). Supe was isolated and the CSP was washed 2 times with 100 mL of sterile saline (pH 7.4) and resuspended in a known volume of sterile saline.
Phospholipid (PL) content of CSP and Supe was measured after organic extraction36 and by quantitation of lipid phosphorus.37 Protein content of CSP and Supe was measured by the bicinchoninic acid methode with bovine serum albumin as the standard. Total phospholipid and protein content was calculated as the sum of their content in CSP and Supe.
PL composition of CSP was determined with high performance liquid chromatography (HPLC) equipped with an evaporative light scatter detector.f Previously described methods of gradient elution were used with a Kromasil silica columng as the solid phase.17,27 A pulsating bubble surfactometer (PBS)h was used to evaluate surface tension lowering activity of CSP at a final standardized concentration of 0.5 mg phospholipid/mL (in buffered saline containing 1.5 mM CaCl2).38 Samples (40 μL) were analyzed at 20 cycles/min for 10 minutes at 37°C in the surfactometer.17,27 Surface tension is reported as γmin, which represents the minimum surface tension achieved over a 10-minute period.
Data are reported as mean ± SD. Normal distribution of the data was evaluated by the Shapiro–Wilk test for normal distribution.39 Data were checked for outliers by scatterplots and standardized residual plots. Associations between variables were first determined by univariate linear regression analysis with PROC REG in SAS.i Independent variables that satisfied P= .1 were introduced into a multivariable regression model by backward and stepwise selection methods. Dependent variables were tested for effect of sampling day by ANOVA by PROC GLM in SAS.
Clinical and BALF Sample Characteristics
Data for age, weight, BALF recovery, cell counts, and cell differentials from 15 horses are presented in Table 1. For 1 of the 17 horses sampled (a 20-year-old mare), the BALF recovery percentage was <50% (42% BALF recovery percentage) and nucleated cell count (NCC) was <5 × 105/mL BALF (2.8 × 105/mL BALF) (see Table 2). Data from this and another horse were excluded from statistical analysis because they were outliers in the data analysis for surfactant characteristics. Data from these 2 horses is presented in Table 2. The neutrophil percentage was <5% in BALF differential cell counts from 11 horses and was between 5 and 15% in the other 4 horses.
Table 1. Clinical and sample characteristics: data are presented for 15 horses as mean ± SD.
Surfactant variables from 15 horses are presented in Table 3. Surfactant variables from the horse with low BALF recovery percentage and low NCC mentioned are presented in Table 4. Surfactant protein content was not determined from this sample. Another horse had a phospholipid content of CSP that was >3 SDs higher than results from the rest of the horses (a 9-year-old gelding). Data from this horse therefore were considered as outlying observations and were excluded from the statistical analysis. Surfactant characteristics from this horse are also presented in Table 4.
Table 3. Surfactant variables: data are presented for 15 horses as mean ± SD.
Concentrations of PE and SPM were not detected on the chromatogram from 1 of the horses.
CSP, crude surfactant pellet; PC, phosphatidylcholine; PC/PG, ratio between PC and PG; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PI, phosphatidylinositol; SPH, sphingomyelin Supe, surfactant supernatant.
Phospholipid (PL) content
PL total (μg/mL)
PL CSP (μg/mL)
PL Supe (μg/mL)
Prot total (μg/mL)
Prot CSP (μg/mL)
Prot Supe (μg/mL)
Surface tension (mN/m)
Table 4. Surfactant variables: individual data for 2 horses excluded from statistical analysis.
CSP, crude surfactant pellet; PC, phosphatidylcholine; PC/PG, ratio between PC and PG; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PI, phosphatidylinositol; SPH, sphingomyelin; Supe, surfactant supernatant.
PL total (μg/mL)
PL CSP (μg/mL)
PL Supe (μg/mL)
Prot total (μg/mL)
Prot CSP (μg/mL)
Prot Supe (μg/mL)
Surface tension (mN/m)
Influence of Age and BALF Sample Characteristics on Surfactant Variables
Increasing age was associated with decreased phospholipid content in CSP but not Supe (Fig 1A). According to the regression equation established from our results, each 1 year increase in age led to a 5.7 μg/mL BALF (95% CI: 1–10 μg/mL BALF) decrease in surfactant phospholipid content. Age did not affect protein content in CSP and Supe (Fig 1B) or phospholipid composition of surfactant (data not presented). An increase in age was not related to significant changes in the ratios of phospholipid in CSP/Supe (r2= 0.1677, P= .1296), protein/phospholipid (r2= 0.1999, P= .0947), and ratio between PC and PG (PC/PG) (r2= 0.0049, P= .8037). Age did not influence BALF recovery percentage (r2= 0.0459, P= .4432), NCC (r2= 0.0495, P= .4253), or cytological profiles (data not presented).
BALF recovery percentage and NCC were unrelated to phospholipid content in CSP (r2= 0.1484, P= .1562 and r2= 0.1272, P= .1920) and Supe (r2= 0.0562, P= .3949 and r2= 0.0448, P= .4489), protein content in CSP (r2= 0.0194, P= .6203 and r2= 0.1486, P= .1559) and Supe (r2= 0.0275, P= .5547 and r2= 0.0055, P= .7921), and phospholipid composition (data not presented). Similarly, BALF recovery percentage and NCC did not significantly influence the ratios CSP/Supe (r2= 0.1677, P= .1296 and r2= 0.0763, P= .3190), protein/phospholipid (r2= 0.1152, P= .2159 and r2= 0.1396, P= .1701), and PC/PG (r2= 0.1767, P= .1188 and r2= 0.0505, P= .4206). Neutrophil percentage of BALF had an effect on the percentage of sphingomyelin (SPH) (SPH%=− 0.0158 × Neutrophil%+ 0.5038, r2= 0.4085, P= .0139). No other effects were found between cell composition of BALF and phospholipid composition of surfactant.
All dependent variables were tested for an effect of sampling day. No effects were detected and sampling day was excluded from further statistical analysis. Multivariable linear regression between phospholipid content in CSP and selected clinical and sample variables only retained age in the model (P= .0144). Multivariable regression between phospholipid content in Supe, protein content in CSP and Supe, and selected clinical and sample parameters did not retain any variables in the model.
Removal or inclusion of the data from 2 horses excluded from statistical data analysis had in most cases only little influence on the coefficients of determination. For example, the coefficient of determination for regression between age and CSP varied from r2= 0.33 to 0.43 depending on the inclusion or exclusion of outliers. However, for the regression analysis between PG% and BALF recovery %, the coefficient of determination changed from r2= 0.18 to = 0.40 when data from the horse with low BALF recovery % and low NCC were removed or included in the analysis. Data from this horse were considered as an outlier based on the substantial change it induced in the coefficient of determination.39
Our study demonstrated the presence of age-related surfactant changes in horses maintained on pasture. More specifically, older horses had lower phospholipid content in CSP, the surfactant-rich fraction of BALF. Increasing age did not affect the ratios of phospholipid in CSP/Supe or protein/phospholipid which may be altered by certain respiratory diseases.17,25 The age-related decrease in surfactant phospholipid content was considered unaffected by variable sample recovery during BALF collection because neither BALF recovery percentage nor BALF NCC had an effect on phospholipid content of surfactant. Low phospholipid content in CSP from older horses was unrelated to any compositional or functional changes of surfactant.
Surfactant changes are well described during pulmonary development in the neonate.40 However, few articles (and none using horses) report age-related surfactant changes in animals beyond the neonatal period.31,35 Similar to our findings, evaluation of surfactant from 38 children without bronchopulmonary disease (ages 3–15 years old) revealed an age-related decrease in surfactant phospholipid content without changes in phospholipid composition or overall protein content of BALF.31 A higher number of alveoli and smaller alveolar size were suggested as possible causes for higher surfactant content in young children. Alveolar number increases rapidly in children during the first year of life and reaches its peak around 10–12 years of age.41 Maximal pulmonary function is achieved around the age of 20–25 years and slowly declines thereafter. Aging is associated with a progressive decrease in alveolar surface area of the lung and enlargement of alveoli.41–44 Alveolar development and age-related changes have not been characterized in the horse. Assuming that aging affects pulmonary structure in a similar way in different animal species, an age-related decrease in surfactant in horses could be explained by a lower number of type II alveolar cells per unit lung volume in older horses.
Another possible explanation for a lower surfactant content in older animals is decreased production, secretion, or recycling of surfactant by type II alveolar cells. A study evaluating macaques between the ages of 1 month and 31 years found a lower number and volume density of lamellar bodies (the intracellular storage form of surfactant) in type II alveolar cells in older animals, indicating age-related changes in surfactant turnover.34 However, in another study, tissue content of saturated PC did not differ with age in lungs collected from human subjects at necropsy.33 Lack of a more thorough knowledge of the effect of age on surfactant composition in humans may be explained by the fact that healthy human subjects do not usually undergo diagnostic procedures such as BALF collection.
In Beagle dogs, lung surfactant from middle-aged (3–7 years old) or old (≥ 12 years old) versus young (3–7 months old) dogs was shown to have a higher proportion of PC and lower proportion of phosphatidylserine (PS), and SPH.35 Comparison of middle-aged to older dogs revealed a lower proportion of PS in the latter, but the functional consequences of this difference are not known. Conversely, in rats, saturated PC content decreased significantly from 1 day to 3 months of age and still decreased between 3 and 29 months.32 Results from these 2 studies indicate that species-related differences may exist in the effect of age on surfactant composition.
Age-related changes in pulmonary function are well characterized in human medicine. Aging of the lung is associated with a reduction in pulmonary elastic recoil (increased compliance) caused by a rearrangement in elastic fibers. Concurrently, chest wall compliance decreases and respiratory muscle strength lessens.41,42,45,46 Alveolar surface exchange area of the lung diminishes secondary to a homogeneous enlargement of air spaces.41,43,44,46 Decreased diameter of small bronchioli and changes in supporting tissue facilitate collapse of smaller airways at low volumes.41,42,45,46 Functional consequences of morphological alterations in the lungs of older subjects include increased residual volume and functional residual capacity, decrease in expiratory flow rate, lower diffusion capacity, and higher ventilation/perfusion imbalance.41,42,45,46 These functional changes do not cause clinical signs in healthy subjects but decrease the reserve capacity of the lung in the presence of disease.41,42 In the horse, age-related changes in pulmonary structure and function are poorly characterized. Horses >20 years old have a lower partial pressure of arterial oxygen, carbon dioxide, and a higher alveolar to arterial pressure gradient compared with 3–8-year-old horses.47 These changes likely are the result of decreased diffusion capacity and increased ventilation perfusion imbalance in the lungs of older horses.
Lower surfactant concentration in older horses has the potential to influence pulmonary function at several levels. Larger alveoli in a more compliant lung may require less surfactant to remain open. Surfactant in healthy animals is present in abundance which offers protection against the effects of surfactant altering or inhibiting agents. However, lower surfactant concentration in older horses may increase their susceptibility to surfactant alterations under disease conditions. Furthermore, low surfactant content in older animals may exacerbate age-related pulmonary changes and alterations in pulmonary function associated with chronic respiratory disease such as recurrent airway obstruction. Less surfactant in bronchioli with decreased diameter and support could lead to an even lower collapse volume.41,42 Diminished mucociliary transport in older patients may be exacerbated by lower surfactant concentrations.45,48 The effect of weakened pulmonary immune defenses in older subjects may be worsened by a decreased barrier against inhaled allergens or pathogens.49,50 To what extent these situations occur in the aging lung is unknown.
The majority of horses sampled for our study were mares (13/15) because mares predominated in the VMRCVM teaching herd. Only a very small number of horses were evaluated for each breed. Because of these limitations, it was not possible to evaluate an effect of breed or sex on surfactant. Healthy horses maintained on pasture typically have <5% neutrophils on BALF cytology.51 Four of the horses sampled for our study had BALF cytology neutrophil percentages >5% but <15%. Most likely this increase in neutrophil percentage was related to exposure of the horses to hay. Several studies report BALF neutrophilia occurring in healthy horses housed in a barn environment and fed hay.52,53
One of the horses originally included in our study had a CSP surfactant phospholipid content approximately twice as high as that of the average for the other horses. This horse was not among the youngest horses and we did not encounter problems with sample collection or analysis. In human medicine, pulmonary alveolar proteinosis is related to an increased accumulation of surfactant in alveoli.1 The horse in our study did not demonstrate any signs of pulmonary disease and likely simply represented individual variation. In general, the total phospholipid content in cell-free BALF was comparable with results obtained from healthy horses in other studies.21,54 PL composition determined by HPLC in our study was characterized by slightly higher percentages of PC and slightly lower percentages of PG compared with results from other groups.21,26,54 These differences in phospholipid composition may be explained by the use of different methodologies. In fact, each study on surfactant in horses has used different techniques to process and analyze samples.21,26,54 Comparison of results among different studies therefore should be done with caution and considering the methods of sample processing and analysis used.1
In this study, surfactant function was evaluated with a PBS and reported as the minimal surface tension of surfactant at a concentration of 0.5 mg phospholipid/mL. By this methodology, surface tension of surfactant from healthy subjects typically reaches values below 2 mN/m17 or approaches zero.27,28 Differences in surfactant surface tension have been described in neonatal foals compared with healthy adult horses by this methodology.27 Factors that alter surface tension include surfactant inhibitors, degradation of surfactant components, or both.1 In the absence of surfactant inhibitors or changes in surfactant composition, an 80% decrease in surfactant concentration is necessary before a significant change in surface tension is seen.55 In this study, the concentration of phospholipid in CSP decreased as horses became older, whereas phospholipid composition and protein concentration were unaltered by age. The discovery that surfactant function was unaffected by animal age suggests that the capacity to produce surfactant is retained in older horses, albeit at a decreased level. The causes for this reduction were not examined in the current study but may include a change in alveolar surface area, a decrease in the number of type II alveolar cells, or their ability to synthesize, secrete, or recycle surfactant.
Exclusion versus inclusion of the data from 1 horse with low BALF recovery and cell count changed the relationship between percentage PG and percentage BALF recovery from weak to moderate. Regression analysis was influenced by this 1 data point which was the only sample with a recovery <50%. Results from a larger number of horses with BALF recovery percentages of <50% are necessary to conclude if percentage PG is related to percentage BALF recovery. Because this data point was excluded from statistical analysis, our discussion and conclusion are only valid for samples with recovery of >50%.
In conclusion, surfactant phospholipid content decreases with age in horses maintained on pasture. BALF sample characteristics did not influence surfactant phospholipid content, composition, or function. Additional studies are needed to evaluate if lower surfactant content in older horses is related to changes in pulmonary function or histology. We speculate that although low surfactant phospholipid content does not lead to detectable clinical disease in the healthy older horse, it may have subclinical effects on performance and may increase respiratory compromise caused by respiratory diseases such as recurrent airway obstruction.
a Detomidine, Pfizer, Exton, PA
b Butorphanol, Fort Dodge, Fort Dodge, IA
c Bivona, Gary, IN
d Casy R, Cell Size Analyzers, CELL Tools Inc, San Francisco, CA
e Pierce, Rockford, IL
f SEDERE, Alfortville, France
g Drachrom, Greensboro, NC
h Electronetics, Amherst, NY
i SAS 9.1.3 Service Pack 4, SAS Institute Inc, Cary, NC
This study was supported in part by the Patricia Bonsall Stuart Fund for Equine Studies (871024). Undine Christmann is supported by a fellowship from the Morris Animal Foundation. The authors thank Barbara Dryman and Bonnie Grier for their technical expertise and support during sample analysis; Joci Forkner, Kerri Cooper, Ryan Gorbutt, and Barbara Kafka for their help with sample collection; Chris Wakley, Kevin Weaver, Pamela Mohr, and the staff of the Veterinary Teaching Hospital at VMRCVM for their assistance and cooperation throughout the project.