Surface Alterations Due to Endotracheal lntubation
ALBERT S. KLAINER, M.D. HERMAN TURNDORF, M.D.
l
WEN-HSIEN WU, M.D.* HRISHI MAEWAL,
M.D.
PHILLIP ALLENDER Morgantown, West Virginia
From the Departments of Medicine (Division of Infectious Diseases) and Anesthesiology, West Virginia University Medical Center, Morgantown, West Virginia 26506. Requests for reprints should be addressed to Dr. Albert S. Klainer, Division of infectious Diseases, West Virginia University Medical Center, Morgantown, West Virginia 26506. Manuscript accepted June 7, 1974. Present address: New York University Medical Center, 560 First Avenue, New York, New York 10016. l
674
May 1975
The varied complications of prolonged endotracheal intubation, including tracheal stenosis, stricture and tracheomalacia, are thought to be directly related to lateral tracheal wall pressure. To investigate the initiation, relevance and duration of these changes, studies were undertaken to assess the effects of intubation on the surface anatomy of the trachea by scanning electron microscopy. Tracheal mucosal morphology was studied in ( 1) anesthetized “normal” dogs; (2) dogs sacrificed 2 hours after intubation with cuff deflated or inflated to the “just seal” point using a cuff system producing the lowest tracheal wall pressure; and (3) dogs intubated for 2 hours and then examined at 2 and 7 days after extubatlon. lntubation without cuff inflation resulted in distinct linear areas of nearly complete ciliary denudation along the tract of tubal insertion within 2 hours. Inflated cuffs produced similar but more widespread changes especially over tracheal rings, indicating that pressure in areas of least resiliency significantly contributes to these alterations. Regeneration of cilia could be seen 2 days after extubation, but many anatomic features remained distorted; at 7 days, regeneration was nearly complete, but isolated areas of denudation could still be identified. Scanning electron micrographs of human tracheas taken at autopsy from patients who had had prolonged intubation with the same cuff system correlated well with those obtained from dogs. The long-term sequelae of prolonged endotracheal intubation and tracheostomy have become a prominent therapeutic problem because of the increased number of patients who survive long-term ventilator therapy. Attention has been focused on the role of endotracheal and tracheostomy tube cuffs as primary etiologic factors in the production of tracheomalacia, tracheoesophageal fistula and tracheal stenosis [l-8]. In an attempt to prevent these complications, new endotracheal and tracheostomy cuffs have been devised, and recommendations have been made favoring the use of certain cuffs which produce the lowest lateral tracheal wall pressure [g-13]. Despite the introduction of a wide variety of low pressure cuffs with varying characteristics and materials, and the precise management of endotracheal and tracheostomy cuffs, complications continue to be seen in as many as 6 per cent of exposed patients [ 131. To elucidate the pathogenesis, relevance and duration of these complications, we report scanning electron microscopy studies of tracheal mucosa exposed to a cuff system demonstrated to produce the lowest lateral wall pressure of any of the presently available cuff systems [9].
The American Journal of Medicine
Volume 58
SURFACE ALTERATIONS
MATERIALS AND METHODS Ten healthy mongrel dogs weighing from 11 to 23 kg and anesthetized with sodium pentobarbital (30 mg/kg intravenously) were studied. Two dogs, acting as controls, were anesthetized then sacrificed immediately with intravenous potassium chloride solution; tracheas were quickly removed, washed and fixed for examination as described herein. In eight dogs a tracheostomy was performed without prior tracheal intubation or manipulation, and a Lanz 28 French (7.0 mm ID) tracheostomy tube (Lanz Medical Products Co., Monroeville, Pennsylvania) was inserted and connected to a heated nebulizer (Puritan-Bennet, PuritanBennet Corporation, Baltimore, Maryland) with a conven-
DUE TO ENDOTRACHEAL
INTUBATION-KLAINER
ET AL.
tional “T” piece. An expiratory limb of 50 ml capacity was added to the expiratory limb of the “T” piece to maintain humidity at the peak inspiratory flow rate. The tracheal cuff was not inflated in two dogs; in two, the cuff was slowly inflated to the “just seal” point while a Bird Mark 8 Ventilator was delivering a tidal volume of 10 ml/kg of body weight. In these dogs a Kulite “flat pak” transducer (Model LQL 125-25, Semiconductor Products, Inc., Ridgefield, New Jersey) was placed on the anterior portion of the cuff to measure and record lateral tracheal wall pressure. In all cases, arterial blood pressure was recorded with a femoral artery catheter connected to a Statham P23 BC transducer and a Grass polygraph 50 recorder, and fluid was infused with a Medicut catheter placed in a femoral vein; systemic arterial blood pressure, arterial
Figufe I. Normal canine trachea. A, view demonstrating the entire surface to be uniform; tracheal rings appear as parallel, concentric, convex bands which project above the surrounding surface. B, view of a representative area demonstrating uniform ciliation. C, view showing uniform ciliation with the cilia oriented in essentially one direction. D, view demonstrating the cilia to be uniform in arrangement with clumps of cilia representing the surfaces of individual cells; occasional goblet cells with microvilli can be seen. Original magnification X 20 (A), X 100 (B), X 500 (C) and X 2,000 (0); reduced by 14 per cent.
May 1975
The American Journal of Medicine
Volume 58
675
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
ET AL.
A
C
Figure 2. Canine trachea after 2 hours of endotracheal intubation with an uninflated cuff. A, view demonstrating areas of linear damage along the tract of tubal insertion. B and C, views showing that areas of linear damage represent interruption of the uniform ciliary mantle. 0, view showing an area of ciliary denudation in which the remaining cilia appear disoriented and heterogeneous in size and shape; the denuded surface is irregular with random interruption of the surface integrity and a suggestion of cilia1 remnants and desquamation. Original magnification X 20 (A), X 50 (S), X 200 (C) and X 2,000 (D); reduced by 14 per cent.
blood gases and pH were
maintained
within normal
during the study. Two hours after tracheostomy or cuff inflation, these four dogs were sacrificed
limits
insertion with intra-
venous saturated potassium chloride solution, and the tracheas were removed for examination. Four additional dogs were extubated after 2 hours of intubation with inflated cuffs and allowed to recover from anesthesia; two each of these dogs were sacrificed 48 hours and 7 days, respectively, after extubation, and the tracheas were removed for examination. In all dogs, the location of the upper border of the tracheal cuff was noted by counting tracheal rings below the stoma site. The tracheostomy cuff was deflated or allowed to remain
676
deflated
May 1975
as the
tracheostomy
The American
tube
Journal of Medicine
was
re-
Volume 56
moved. The trachea moved
by clean
below the stoma was then quickly re-
dissection,
solution and immediately specimens were examined at low magnification, and
washed
with warm
fixed as described
The
with a dissecting microscope characteristic regions at and
below the cuff site were selected
for study at higher mag-
nification with the scanning electron microscope bridge Stereoscan S4- 10, Cambridge Scientific ments Limited, Cambridge,
Ringer’s
herein.
(CamInstru-
England).
The tracheas of four patients who died of nonairway disease were removed and fixed for scanning electron microscope examination within 90 minutes of the time of death. Two patients had not had an endotracheal or tracheostomy tube in place prior to death nor had their air-
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
ways been manipulated. Two patients died 8 and 14 days after tracheostomy with the Lanz system tracheostomy cuff in place. Meticulous attention had been paid to the details of tracheostomy management and cuff care by nurses trained in intensive care throughout the entire course of illness in these patients. For scanning electron microscope examination, trachea specimens were transected and opened by longitudinal incision along the dorsal surface. The mucous membrane was immediately washed with warm Ringer’s solution to remove mucus, debris and red blood cells. A 20 cc syringe with a plastic nozzle (tip ID of 0.5 mm) was used to obtain a gentle, but thorough, washing action on the mucosal surface. Appropriate representative sections were carefully removed for preparation for scanning electron microscope examination. Sectioned specimens were immediately placed in 2 per cent cacodylate-buffered (pH 7.3) gtuteraldehyde and fixed overnight. Fixed sections were then washed in distilled
ET AL.
water to remove gluteraldehyde; dehydrated in a graded series of 20, 50, 75, 90 and 100 per cent ethanol; and isoamyl acetate was substituted for ethanol in concentrations of 35, 50 and 100 per cent in preparation for drying with liquid carbon dioxide in the Denton DCP-1, Critical Point Drying Apparatus (Denton Vacuum, Inc.. Cherry Hill, N.J.). Dried specimens were mounted on aluminum stubs with conducting silver paint, coated with 100 to 150 A carbon and gold palladium alloy, and examined in a Cambridge S4 Stereoscan scanning electron microscope at 20 KV with a tilt angle of 35 degrees. Sections of representative tissue were also fixed and stained with hematoxylin and eosin for conventional histologic examination by light microscopy. RESULTS The surface morphology of the trachea of control dogs as viewed by scanning electron microscopy of
Figure 3. A composite view of the lateral wall of canine trachea 2 hours after endotracheal intubation with the cuff inflated to “‘just seal” point. The lighter areas demonstrate relatively good preservation of ciliary integrity. The darker areas, predominantly overlying the tracheal rings and underlying the cuff or distal end of the tube, show ciliary denudation. Original magnification X 20, reduced by 44 per cent.
May 1975
The American Journal of Medicine
Volume 58
677
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
ET AL.
Figure 4. Canine trachea (lateral wall) 2 hours after endotracheal intubation with the cuff inflated to “just seal” point. A, the insert represents the area shown at higher magnifications in B, C and D. B, view showing transition zone with the light area corresponding to the dark area seen at lower magnification due to electron beam interaction. C, view showing denuded area (lower field) in which there are only irregular clumps of cilia, occasional red blood cells and larger spherical cells consistent with inflammatory cells; in the upper field, cilia are matted together. D, view similar to that shown in C. Original magnification X 20 (A), X 100 (B), X 500 (C) and X 2,000 (0); reduced by 14 per cent.
selected representative fields of increasing magnification is shown in Figure 1. The entire surface of the trachea is uniformly covered with dense cilia punctuated only by the appearance of occasional goblet cells. At low magnification, cilia are consistently present on the mucosal surface over and between tracheal rings; tracheal rings appear as parallel, convex bands which project above the surrounding surface. The cilia are uniform in arrangement, and clumps representing the surfaces of individual cells are visible; cilia are oriented in essentially one direc-
676
May 1975
The American Journal of Medicine
Volume 66
tion and are monotonously uniform in shape, length and diameter. Figure 2 demonstrates the surface morphology of the trachea of a dog intubated with an uncuffed tube for 2 hours. At low magnification, the uniformity of the cilia1 mantle is interrupted only by linear and occasional transverse areas over which cilia are lost. At increasing magnifications, these areas are shown to be characterized by ciliary denudation in which only occasional cells appear ciliated: cilia are disoriented and appear less homogeneous in size,
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
shape and diameter, and resemble irregular tufts. The denuded surface is irregular and coarse with random interruption of its integrity and a suggestion of cilia1 remnants and desquamation. Figures, 3 through 5 illustrate surface changes in dogs intubated for 2 hours with tubes cuffed to “just seal” point. A low magnification composite view (Figure 3) demonstrates ciliary denudation visible predominantly over the tracheal rings in contact with the upper margin of the tube: relatively good preservation of morphology remains in the intercartilagenous areas. The trachea lying beneath the distal portion of the tube and the cuff is almost completely denuded
F&we 5. Canine trachea (lateral point. A and 6, views of transitional denudation in the area of intubation. integrity. 0, view of an intact area 5,000 (C and 0); reduced by 14 per
ET AL.
of cilia. The most diffuse area of ciliary denudation is seen underlying the cuff site (Figures 4 and 5). In addition, red blood cells and large spherical cells consistent with inflammatory cells are present (Figures 4C, 4D and 58). A transitional zone bordering normal and abnormal areas can be identified in which the cilia appear matted without uniformity of direction (Figures 4B, 4C, 5A and 5B) as compared to more normal areas (Figure 5D). The surfaces of the denuded areas vary from irregular with ciliary remnants (Figures 4C and 4D) to smooth with almost complete loss of morphologic integrity (Figure 5C). Because the scanning electron microscope inter-
wall} 2 hours after endotracheal intubation with the cuff inflated to “just seal” zone demonstrating matting together of the cilia where ciliation is preserved and C, view demonstrating denuded area with almost complete loss of morphologic showing normal ciliation. Original magnification X 700 (A), X 7,000 (B) and X cent.
May 1975
The American Journal of Medicine
Volume 56
679
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
ET AL.
Figure 6. Representative views of standard hematoxylin and eosin stained sections of canine trachea after 2 hours’ intubation with cuff inflated to “just seal” point. A, normal ciliated tracheal mucosal ceils. B, trachea/ mucosal cells with loss of cilia due to intubation. C, transition zone between intact and damaged area. To the riuht. intact ciliated tracheal mucosal cells; to the left, damaged area along the tract of tubal inserti& showing almost Ebkplete destruction of the mucosa. D, area under the cuff showing almost complete Ioss of tracheal mucosa. Original magnification X 500; reduced by 30 per cent.
pretation of tracheal morphology demands correlation between what is seen at high magnification in three-dimensional perspective and what has been readily accepted as histologic structure viewed by standard light microscopy of thin sections, representative areas were examined after hematoxylin and eosin staining. Figure 6 shows that when appropriately ChDSen areas are examined by light microscopy, changes almost identical to those observed by scanning microscopy can be demonstrated. Although identical sections were not examined by both forms of microscopy, carefully selected portions of tissue representing the same anatomic and pathologic areas were examined. The sections shown in Figure 6 were specifically chosen from areas comparable to the transitional zone between normal and injured trachea shown in Figures 2. 4 and 5. Although light microscopy cannot provide the depth of field, the magnification or the three-dimensional perspective of scanning microscopy, it does provide information about the changes in epithelial cells in cross-section. Had the trachea not been inspected by scanning microscopy, it is unlikely that the zones of change would have been identified since gross surface appearance usually did not differ significantly from normal. Examination of canine tracheas obtained 48 hours and 7 days after extubation (Figure 7) revealed early
680
May 1975
The Amerlcafi Journal of Medicine
Volume 58
regeneration of cilia by 48 hours (Figures 7A, 7B and 7C); cilia of varying size could be seen, but these were generally short and rudimentary; when longer cilia could be observed, they were frequently matted together and in a state of disarray. By 7 days, ciliation was nearly complete, but isolated denuded areas could still be identified (Figure 7D). To evaluate the possible significance of tube-induced injury, we studied autopsy material from four patients, two of whom had not been exposed to endotracheal intubation and two of whom had been intubated with the same cuff system utilized in dogs for 8 and 14 days, respectively. In unintubated patients (Figure 8A), mucosal morphology was well preserved and revealed uniform ciliation with occasional goblet cells; cilia, once again, were oriented in one direction and appeared to be uniform in size, shape and diameter. In contrast, the tracheas of intubated patients (Figure 8B) revealed alterations similar to those observed in experimental dogs. Areas of injury were devoid of cilia with changes suggesting squamous metaplasia characterized by flat, elongated, hexagonal cells with granular surfaces representing either microvilli or ciliary remnants. COMMENTS This study provides convincing evidence that cuffed endotracheal and tracheostomy tubes produce mea-
SURFACE ALTERATIONS
surable tracheal injury even if meticulous attention is given to the selection of the cuff system and the method of inflation. Since the morphologic alterations were limited to the region of the tube and cuff, it would appear that injury is the direct result of pressure on the tracheal wall rather than the effects of inadequate humidity or other variables. Ciliary denudation at points of contact between the tube and trachea, where contact pressure is presumably extremely low, has not been describecl before. It is surprising that lateral wall pressures as low as 18 to 25 mm Hg could produce these lesions in normotensive animals especially in as short ,a period as 2 hours. In addition, it is of im-
DUE TO ENDOTRACHEAL
INTUBATION-KLAINER
ET AL.
mense clinical significance that changes observed in dogs correlated well with those observed in human subjects. The preliminary regeneration studies described suggest that, following periods of intubation as short as 2 hours, cilia do regenerate by 2 to 7 days but that total anatomic integrity is not accomplished. It is obvious that these morphologic observations must be complemented by functional studies to correlate structural damage with possible physiologic dysfunction. Whether ciliary regeneration will occur after longer periods of low pressure cuff inflation remains to be answered. If these changes
have
any long-lasting
anatomic
Figure 7. A, B and C, canine trachea 48 hours after extubation. A, an area of early regeneration with rudimentary cilia appearing in selected areas. B, area showing regeneration of early rudimentary cilia. C, area showing matting together of individual cilia. 0, canine trachea 7 days after extubation showing a/most total regeneration of cilia but still discrete areas of denudation. Original magnification X 800 (A and D), X 5,000 (B) and X 15,000 (C); reduced by 14 per cent.
May 1975
The American Journal of Medicine
Volume 68
661
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
ET AL.
Figure 8. A, normal human trachea. B, trachea from patient 8 days after endotracheal intubation; note almost complete loss of cilia and tendency toward squamous metaplasia. Original magnification X 2,500.
682
May 1975
The American
Journal of Medicine
Volume 58
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
significance, the concept that low-pressure cuffs should be made broader [13] is vitiated since the area of cuff-induced injury would thereby be extended. These studies also suggest that still more emphasis must be placed on cuff management during brief periods of intubation such as frequently used for the administration of anesthesia. Although it is apparent that low pressure cuffs may produce less tissue injury than even well managed high pressure cuffs, it is also apparent that injury nevertheless occurs. Low pressure cuffs are not yet the final solution to the tracheal injury problem. Obviously no recommendation to avoid endotracheal
intubation can be made without many other factors.
ET AL.
consideration
of
ACKNOWLEDGMENTS We are grateful to Susanna Jernigan, Division of Infectious Diseases and Marylon Rahber and Fred Simpson, Department of Anesthesiology, West Virginia University Medical Center, Morgantown, West Virginia and to Patsy Willard of the Appalachian Laboratory for Occupational and Respiratory Diseases, Morgantown, West Virginia for their technical assistance.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
Andrews MJ, Pearson FG: Incidence and pathogenesis of tracheal injury following cuffed tube tracheostomy with assisted ventilation: analysis of a two-year prospective study. Ann Surg 173: 249, 1971. Geffin B, Grill0 HC, Cooper JD. Pontoppidan H: Stenosis following tracheostomy for respiratory care. JAMA 216: 1984, 1971. Rainer WG, Sanchez M, Lopez L: Tracheal stricture secondary to cuffed tracheostomy tubes. Chest 59: 115, 1971. Cooper JD, Grill0 HC: The evolution of tracheal injury due to ventilatory assistance through cuffed tubes: a pathologic study. Ann Surg 169: 334, 1969. Cooper JD, Grill0 HC: Experimental production and prevention of injury due to cuffed tracheal tubes. Surg Gynecol Olbstet 129: 1235, 1969. Shelly WM. Dawson RB, May IA: Cuffed tubes as a cause oi tracheal stenosis. J Thorac Cardiovasc Surg 57: 623, 1969. Brya.nt LR, Trinkle JK, Dubilier L: Reappraisal of tracheal in-
8.
9.
10. 11.
12.
13.
May 1975
jury from cuffed tracheostomy tubes: experiments in dogs. JAMA 215: 625, 1971. Hedden M, Ersoz CJ, Safar P: Tracheoesophageal fistulas following prolonged artificial ventilation via cuffed tracheostomy tubes. Anesthesiology 31: 281, 1969. Wu WH, Lim IT, Simpson FA, Turndorf H: Pressure dynamics of endotracheal and tracheostomy cuffs. I. Use of a tracheal model to evaluate performance. Crit Care Med 1: 197, 1973. Nealon TF, Ching N: Pressures of tracheostomy cuffs in ventilated patients. NY State J Med 71: 1923, 1971. Knowlson GTG, Bassett HFM: The pressures exerted on the trachea by endotracheal inflatable cuffs Br J Anaesth 42: 834, 1970. Grill0 HC, Cooper JD, Geffin B, Pontoppidan H: A low-pressure cuff for tracheostomy tubes to minimize tracheal injury. J Thorac Cardiovasc Surg 62: 898, 1971. Geffin B, Pontoppidan H: Reduction of tracheal damage by the prestretching of inflatable cuffs. Anesthesiology 31: 462, 1969.
The American Journal of Mediclne
Volume 58
683
ALBERT S. KLAINER, M.D. HERMAN TURNDORF, M.D.
l
WEN-HSIEN WU, M.D.* HRISHI MAEWAL,
M.D.
PHILLIP ALLENDER Morgantown, West Virginia
From the Departments of Medicine (Division of Infectious Diseases) and Anesthesiology, West Virginia University Medical Center, Morgantown, West Virginia 26506. Requests for reprints should be addressed to Dr. Albert S. Klainer, Division of infectious Diseases, West Virginia University Medical Center, Morgantown, West Virginia 26506. Manuscript accepted June 7, 1974. Present address: New York University Medical Center, 560 First Avenue, New York, New York 10016. l
674
May 1975
The varied complications of prolonged endotracheal intubation, including tracheal stenosis, stricture and tracheomalacia, are thought to be directly related to lateral tracheal wall pressure. To investigate the initiation, relevance and duration of these changes, studies were undertaken to assess the effects of intubation on the surface anatomy of the trachea by scanning electron microscopy. Tracheal mucosal morphology was studied in ( 1) anesthetized “normal” dogs; (2) dogs sacrificed 2 hours after intubation with cuff deflated or inflated to the “just seal” point using a cuff system producing the lowest tracheal wall pressure; and (3) dogs intubated for 2 hours and then examined at 2 and 7 days after extubatlon. lntubation without cuff inflation resulted in distinct linear areas of nearly complete ciliary denudation along the tract of tubal insertion within 2 hours. Inflated cuffs produced similar but more widespread changes especially over tracheal rings, indicating that pressure in areas of least resiliency significantly contributes to these alterations. Regeneration of cilia could be seen 2 days after extubation, but many anatomic features remained distorted; at 7 days, regeneration was nearly complete, but isolated areas of denudation could still be identified. Scanning electron micrographs of human tracheas taken at autopsy from patients who had had prolonged intubation with the same cuff system correlated well with those obtained from dogs. The long-term sequelae of prolonged endotracheal intubation and tracheostomy have become a prominent therapeutic problem because of the increased number of patients who survive long-term ventilator therapy. Attention has been focused on the role of endotracheal and tracheostomy tube cuffs as primary etiologic factors in the production of tracheomalacia, tracheoesophageal fistula and tracheal stenosis [l-8]. In an attempt to prevent these complications, new endotracheal and tracheostomy cuffs have been devised, and recommendations have been made favoring the use of certain cuffs which produce the lowest lateral tracheal wall pressure [g-13]. Despite the introduction of a wide variety of low pressure cuffs with varying characteristics and materials, and the precise management of endotracheal and tracheostomy cuffs, complications continue to be seen in as many as 6 per cent of exposed patients [ 131. To elucidate the pathogenesis, relevance and duration of these complications, we report scanning electron microscopy studies of tracheal mucosa exposed to a cuff system demonstrated to produce the lowest lateral wall pressure of any of the presently available cuff systems [9].
The American Journal of Medicine
Volume 58
SURFACE ALTERATIONS
MATERIALS AND METHODS Ten healthy mongrel dogs weighing from 11 to 23 kg and anesthetized with sodium pentobarbital (30 mg/kg intravenously) were studied. Two dogs, acting as controls, were anesthetized then sacrificed immediately with intravenous potassium chloride solution; tracheas were quickly removed, washed and fixed for examination as described herein. In eight dogs a tracheostomy was performed without prior tracheal intubation or manipulation, and a Lanz 28 French (7.0 mm ID) tracheostomy tube (Lanz Medical Products Co., Monroeville, Pennsylvania) was inserted and connected to a heated nebulizer (Puritan-Bennet, PuritanBennet Corporation, Baltimore, Maryland) with a conven-
DUE TO ENDOTRACHEAL
INTUBATION-KLAINER
ET AL.
tional “T” piece. An expiratory limb of 50 ml capacity was added to the expiratory limb of the “T” piece to maintain humidity at the peak inspiratory flow rate. The tracheal cuff was not inflated in two dogs; in two, the cuff was slowly inflated to the “just seal” point while a Bird Mark 8 Ventilator was delivering a tidal volume of 10 ml/kg of body weight. In these dogs a Kulite “flat pak” transducer (Model LQL 125-25, Semiconductor Products, Inc., Ridgefield, New Jersey) was placed on the anterior portion of the cuff to measure and record lateral tracheal wall pressure. In all cases, arterial blood pressure was recorded with a femoral artery catheter connected to a Statham P23 BC transducer and a Grass polygraph 50 recorder, and fluid was infused with a Medicut catheter placed in a femoral vein; systemic arterial blood pressure, arterial
Figufe I. Normal canine trachea. A, view demonstrating the entire surface to be uniform; tracheal rings appear as parallel, concentric, convex bands which project above the surrounding surface. B, view of a representative area demonstrating uniform ciliation. C, view showing uniform ciliation with the cilia oriented in essentially one direction. D, view demonstrating the cilia to be uniform in arrangement with clumps of cilia representing the surfaces of individual cells; occasional goblet cells with microvilli can be seen. Original magnification X 20 (A), X 100 (B), X 500 (C) and X 2,000 (0); reduced by 14 per cent.
May 1975
The American Journal of Medicine
Volume 58
675
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
ET AL.
A
C
Figure 2. Canine trachea after 2 hours of endotracheal intubation with an uninflated cuff. A, view demonstrating areas of linear damage along the tract of tubal insertion. B and C, views showing that areas of linear damage represent interruption of the uniform ciliary mantle. 0, view showing an area of ciliary denudation in which the remaining cilia appear disoriented and heterogeneous in size and shape; the denuded surface is irregular with random interruption of the surface integrity and a suggestion of cilia1 remnants and desquamation. Original magnification X 20 (A), X 50 (S), X 200 (C) and X 2,000 (D); reduced by 14 per cent.
blood gases and pH were
maintained
within normal
during the study. Two hours after tracheostomy or cuff inflation, these four dogs were sacrificed
limits
insertion with intra-
venous saturated potassium chloride solution, and the tracheas were removed for examination. Four additional dogs were extubated after 2 hours of intubation with inflated cuffs and allowed to recover from anesthesia; two each of these dogs were sacrificed 48 hours and 7 days, respectively, after extubation, and the tracheas were removed for examination. In all dogs, the location of the upper border of the tracheal cuff was noted by counting tracheal rings below the stoma site. The tracheostomy cuff was deflated or allowed to remain
676
deflated
May 1975
as the
tracheostomy
The American
tube
Journal of Medicine
was
re-
Volume 56
moved. The trachea moved
by clean
below the stoma was then quickly re-
dissection,
solution and immediately specimens were examined at low magnification, and
washed
with warm
fixed as described
The
with a dissecting microscope characteristic regions at and
below the cuff site were selected
for study at higher mag-
nification with the scanning electron microscope bridge Stereoscan S4- 10, Cambridge Scientific ments Limited, Cambridge,
Ringer’s
herein.
(CamInstru-
England).
The tracheas of four patients who died of nonairway disease were removed and fixed for scanning electron microscope examination within 90 minutes of the time of death. Two patients had not had an endotracheal or tracheostomy tube in place prior to death nor had their air-
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
ways been manipulated. Two patients died 8 and 14 days after tracheostomy with the Lanz system tracheostomy cuff in place. Meticulous attention had been paid to the details of tracheostomy management and cuff care by nurses trained in intensive care throughout the entire course of illness in these patients. For scanning electron microscope examination, trachea specimens were transected and opened by longitudinal incision along the dorsal surface. The mucous membrane was immediately washed with warm Ringer’s solution to remove mucus, debris and red blood cells. A 20 cc syringe with a plastic nozzle (tip ID of 0.5 mm) was used to obtain a gentle, but thorough, washing action on the mucosal surface. Appropriate representative sections were carefully removed for preparation for scanning electron microscope examination. Sectioned specimens were immediately placed in 2 per cent cacodylate-buffered (pH 7.3) gtuteraldehyde and fixed overnight. Fixed sections were then washed in distilled
ET AL.
water to remove gluteraldehyde; dehydrated in a graded series of 20, 50, 75, 90 and 100 per cent ethanol; and isoamyl acetate was substituted for ethanol in concentrations of 35, 50 and 100 per cent in preparation for drying with liquid carbon dioxide in the Denton DCP-1, Critical Point Drying Apparatus (Denton Vacuum, Inc.. Cherry Hill, N.J.). Dried specimens were mounted on aluminum stubs with conducting silver paint, coated with 100 to 150 A carbon and gold palladium alloy, and examined in a Cambridge S4 Stereoscan scanning electron microscope at 20 KV with a tilt angle of 35 degrees. Sections of representative tissue were also fixed and stained with hematoxylin and eosin for conventional histologic examination by light microscopy. RESULTS The surface morphology of the trachea of control dogs as viewed by scanning electron microscopy of
Figure 3. A composite view of the lateral wall of canine trachea 2 hours after endotracheal intubation with the cuff inflated to “‘just seal” point. The lighter areas demonstrate relatively good preservation of ciliary integrity. The darker areas, predominantly overlying the tracheal rings and underlying the cuff or distal end of the tube, show ciliary denudation. Original magnification X 20, reduced by 44 per cent.
May 1975
The American Journal of Medicine
Volume 58
677
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
ET AL.
Figure 4. Canine trachea (lateral wall) 2 hours after endotracheal intubation with the cuff inflated to “just seal” point. A, the insert represents the area shown at higher magnifications in B, C and D. B, view showing transition zone with the light area corresponding to the dark area seen at lower magnification due to electron beam interaction. C, view showing denuded area (lower field) in which there are only irregular clumps of cilia, occasional red blood cells and larger spherical cells consistent with inflammatory cells; in the upper field, cilia are matted together. D, view similar to that shown in C. Original magnification X 20 (A), X 100 (B), X 500 (C) and X 2,000 (0); reduced by 14 per cent.
selected representative fields of increasing magnification is shown in Figure 1. The entire surface of the trachea is uniformly covered with dense cilia punctuated only by the appearance of occasional goblet cells. At low magnification, cilia are consistently present on the mucosal surface over and between tracheal rings; tracheal rings appear as parallel, convex bands which project above the surrounding surface. The cilia are uniform in arrangement, and clumps representing the surfaces of individual cells are visible; cilia are oriented in essentially one direc-
676
May 1975
The American Journal of Medicine
Volume 66
tion and are monotonously uniform in shape, length and diameter. Figure 2 demonstrates the surface morphology of the trachea of a dog intubated with an uncuffed tube for 2 hours. At low magnification, the uniformity of the cilia1 mantle is interrupted only by linear and occasional transverse areas over which cilia are lost. At increasing magnifications, these areas are shown to be characterized by ciliary denudation in which only occasional cells appear ciliated: cilia are disoriented and appear less homogeneous in size,
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
shape and diameter, and resemble irregular tufts. The denuded surface is irregular and coarse with random interruption of its integrity and a suggestion of cilia1 remnants and desquamation. Figures, 3 through 5 illustrate surface changes in dogs intubated for 2 hours with tubes cuffed to “just seal” point. A low magnification composite view (Figure 3) demonstrates ciliary denudation visible predominantly over the tracheal rings in contact with the upper margin of the tube: relatively good preservation of morphology remains in the intercartilagenous areas. The trachea lying beneath the distal portion of the tube and the cuff is almost completely denuded
F&we 5. Canine trachea (lateral point. A and 6, views of transitional denudation in the area of intubation. integrity. 0, view of an intact area 5,000 (C and 0); reduced by 14 per
ET AL.
of cilia. The most diffuse area of ciliary denudation is seen underlying the cuff site (Figures 4 and 5). In addition, red blood cells and large spherical cells consistent with inflammatory cells are present (Figures 4C, 4D and 58). A transitional zone bordering normal and abnormal areas can be identified in which the cilia appear matted without uniformity of direction (Figures 4B, 4C, 5A and 5B) as compared to more normal areas (Figure 5D). The surfaces of the denuded areas vary from irregular with ciliary remnants (Figures 4C and 4D) to smooth with almost complete loss of morphologic integrity (Figure 5C). Because the scanning electron microscope inter-
wall} 2 hours after endotracheal intubation with the cuff inflated to “just seal” zone demonstrating matting together of the cilia where ciliation is preserved and C, view demonstrating denuded area with almost complete loss of morphologic showing normal ciliation. Original magnification X 700 (A), X 7,000 (B) and X cent.
May 1975
The American Journal of Medicine
Volume 56
679
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
ET AL.
Figure 6. Representative views of standard hematoxylin and eosin stained sections of canine trachea after 2 hours’ intubation with cuff inflated to “just seal” point. A, normal ciliated tracheal mucosal ceils. B, trachea/ mucosal cells with loss of cilia due to intubation. C, transition zone between intact and damaged area. To the riuht. intact ciliated tracheal mucosal cells; to the left, damaged area along the tract of tubal inserti& showing almost Ebkplete destruction of the mucosa. D, area under the cuff showing almost complete Ioss of tracheal mucosa. Original magnification X 500; reduced by 30 per cent.
pretation of tracheal morphology demands correlation between what is seen at high magnification in three-dimensional perspective and what has been readily accepted as histologic structure viewed by standard light microscopy of thin sections, representative areas were examined after hematoxylin and eosin staining. Figure 6 shows that when appropriately ChDSen areas are examined by light microscopy, changes almost identical to those observed by scanning microscopy can be demonstrated. Although identical sections were not examined by both forms of microscopy, carefully selected portions of tissue representing the same anatomic and pathologic areas were examined. The sections shown in Figure 6 were specifically chosen from areas comparable to the transitional zone between normal and injured trachea shown in Figures 2. 4 and 5. Although light microscopy cannot provide the depth of field, the magnification or the three-dimensional perspective of scanning microscopy, it does provide information about the changes in epithelial cells in cross-section. Had the trachea not been inspected by scanning microscopy, it is unlikely that the zones of change would have been identified since gross surface appearance usually did not differ significantly from normal. Examination of canine tracheas obtained 48 hours and 7 days after extubation (Figure 7) revealed early
680
May 1975
The Amerlcafi Journal of Medicine
Volume 58
regeneration of cilia by 48 hours (Figures 7A, 7B and 7C); cilia of varying size could be seen, but these were generally short and rudimentary; when longer cilia could be observed, they were frequently matted together and in a state of disarray. By 7 days, ciliation was nearly complete, but isolated denuded areas could still be identified (Figure 7D). To evaluate the possible significance of tube-induced injury, we studied autopsy material from four patients, two of whom had not been exposed to endotracheal intubation and two of whom had been intubated with the same cuff system utilized in dogs for 8 and 14 days, respectively. In unintubated patients (Figure 8A), mucosal morphology was well preserved and revealed uniform ciliation with occasional goblet cells; cilia, once again, were oriented in one direction and appeared to be uniform in size, shape and diameter. In contrast, the tracheas of intubated patients (Figure 8B) revealed alterations similar to those observed in experimental dogs. Areas of injury were devoid of cilia with changes suggesting squamous metaplasia characterized by flat, elongated, hexagonal cells with granular surfaces representing either microvilli or ciliary remnants. COMMENTS This study provides convincing evidence that cuffed endotracheal and tracheostomy tubes produce mea-
SURFACE ALTERATIONS
surable tracheal injury even if meticulous attention is given to the selection of the cuff system and the method of inflation. Since the morphologic alterations were limited to the region of the tube and cuff, it would appear that injury is the direct result of pressure on the tracheal wall rather than the effects of inadequate humidity or other variables. Ciliary denudation at points of contact between the tube and trachea, where contact pressure is presumably extremely low, has not been describecl before. It is surprising that lateral wall pressures as low as 18 to 25 mm Hg could produce these lesions in normotensive animals especially in as short ,a period as 2 hours. In addition, it is of im-
DUE TO ENDOTRACHEAL
INTUBATION-KLAINER
ET AL.
mense clinical significance that changes observed in dogs correlated well with those observed in human subjects. The preliminary regeneration studies described suggest that, following periods of intubation as short as 2 hours, cilia do regenerate by 2 to 7 days but that total anatomic integrity is not accomplished. It is obvious that these morphologic observations must be complemented by functional studies to correlate structural damage with possible physiologic dysfunction. Whether ciliary regeneration will occur after longer periods of low pressure cuff inflation remains to be answered. If these changes
have
any long-lasting
anatomic
Figure 7. A, B and C, canine trachea 48 hours after extubation. A, an area of early regeneration with rudimentary cilia appearing in selected areas. B, area showing regeneration of early rudimentary cilia. C, area showing matting together of individual cilia. 0, canine trachea 7 days after extubation showing a/most total regeneration of cilia but still discrete areas of denudation. Original magnification X 800 (A and D), X 5,000 (B) and X 15,000 (C); reduced by 14 per cent.
May 1975
The American Journal of Medicine
Volume 68
661
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
ET AL.
Figure 8. A, normal human trachea. B, trachea from patient 8 days after endotracheal intubation; note almost complete loss of cilia and tendency toward squamous metaplasia. Original magnification X 2,500.
682
May 1975
The American
Journal of Medicine
Volume 58
SURFACE ALTERATIONS DUE TO ENDOTRACHEAL INTUBATION-KLAINER
significance, the concept that low-pressure cuffs should be made broader [13] is vitiated since the area of cuff-induced injury would thereby be extended. These studies also suggest that still more emphasis must be placed on cuff management during brief periods of intubation such as frequently used for the administration of anesthesia. Although it is apparent that low pressure cuffs may produce less tissue injury than even well managed high pressure cuffs, it is also apparent that injury nevertheless occurs. Low pressure cuffs are not yet the final solution to the tracheal injury problem. Obviously no recommendation to avoid endotracheal
intubation can be made without many other factors.
ET AL.
consideration
of
ACKNOWLEDGMENTS We are grateful to Susanna Jernigan, Division of Infectious Diseases and Marylon Rahber and Fred Simpson, Department of Anesthesiology, West Virginia University Medical Center, Morgantown, West Virginia and to Patsy Willard of the Appalachian Laboratory for Occupational and Respiratory Diseases, Morgantown, West Virginia for their technical assistance.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
Andrews MJ, Pearson FG: Incidence and pathogenesis of tracheal injury following cuffed tube tracheostomy with assisted ventilation: analysis of a two-year prospective study. Ann Surg 173: 249, 1971. Geffin B, Grill0 HC, Cooper JD. Pontoppidan H: Stenosis following tracheostomy for respiratory care. JAMA 216: 1984, 1971. Rainer WG, Sanchez M, Lopez L: Tracheal stricture secondary to cuffed tracheostomy tubes. Chest 59: 115, 1971. Cooper JD, Grill0 HC: The evolution of tracheal injury due to ventilatory assistance through cuffed tubes: a pathologic study. Ann Surg 169: 334, 1969. Cooper JD, Grill0 HC: Experimental production and prevention of injury due to cuffed tracheal tubes. Surg Gynecol Olbstet 129: 1235, 1969. Shelly WM. Dawson RB, May IA: Cuffed tubes as a cause oi tracheal stenosis. J Thorac Cardiovasc Surg 57: 623, 1969. Brya.nt LR, Trinkle JK, Dubilier L: Reappraisal of tracheal in-
8.
9.
10. 11.
12.
13.
May 1975
jury from cuffed tracheostomy tubes: experiments in dogs. JAMA 215: 625, 1971. Hedden M, Ersoz CJ, Safar P: Tracheoesophageal fistulas following prolonged artificial ventilation via cuffed tracheostomy tubes. Anesthesiology 31: 281, 1969. Wu WH, Lim IT, Simpson FA, Turndorf H: Pressure dynamics of endotracheal and tracheostomy cuffs. I. Use of a tracheal model to evaluate performance. Crit Care Med 1: 197, 1973. Nealon TF, Ching N: Pressures of tracheostomy cuffs in ventilated patients. NY State J Med 71: 1923, 1971. Knowlson GTG, Bassett HFM: The pressures exerted on the trachea by endotracheal inflatable cuffs Br J Anaesth 42: 834, 1970. Grill0 HC, Cooper JD, Geffin B, Pontoppidan H: A low-pressure cuff for tracheostomy tubes to minimize tracheal injury. J Thorac Cardiovasc Surg 62: 898, 1971. Geffin B, Pontoppidan H: Reduction of tracheal damage by the prestretching of inflatable cuffs. Anesthesiology 31: 462, 1969.
The American Journal of Mediclne
Volume 58
683
