EVALUATION OF DIAPHRAGMATIC MOTION IN NORMAL AND

DIAPHRAGMATIC PARALYZED DOGS USING M-MODE

ULTRASONOGRAPHY

MIHYUN CHOI, NAMSOON LEE, AHYOUNG KIM, SEOYEON KEH, JINSOO LEE, HYUNWOOK KIM, MINCHEOL CHOI

Diagnosis of unilateral diaphragmatic paralysis in dogs is currently based on fluoroscopic detection of unequal
movement between the crura. Bilateral paralysis may be more difficult to confirm with fluoroscopy because
diaphragmatic movement is sometimes produced by compensatory abdominal muscle contractions. The purpose
of this study was to develop a new method to evaluate diaphragmatic movement using M-mode ultrasonography
and to describe findings for normal and diaphragmatic paralyzed dogs. Fifty-five clinically normal dogs and
two dogs with diaphragmatic paralysis were recruited. Thoracic radiographs were acquired for all dogs and
fluoroscopy studies were also acquired for clinically affected dogs. Two observers independently measured
diaphragmatic direction of motion and amplitude of excursion using M-mode ultrasonography for dogs meeting
study inclusion criteria. Eight of the clinically normal dogs were excluded due to abnormal thoracic radiographic
findings. For the remaining normal dogs, the lower limit values of diaphragmatic excursion were 2.85–2.98
mm during normal breathing. One dog with bilateral diaphragmatic paralysis showed paradoxical movement
of both crura at the end of inspiration. One dog with unilateral diaphragmatic paralysis had diaphragmatic
excursion values of 2.00 ± 0.42 mm on the left side and 4.05 ± 1.48 mm on the right side. The difference between
left and right diaphragmatic excursion values was 55%. Findings indicated that M-mode ultrasonography is
a relatively simple and objective method for measuring diaphragmatic movement in dogs. Future studies are
needed in a larger number of dogs with diaphragmatic paralysis to determine the diagnostic sensitivity of this
promising new technique.  C 2013 American College of Veterinary Radiology.

Key words: diaphragmatic paralysis, dog, M-mode ultrasonography.

Introduction muscle or accessory respiratory muscles leading to respira-

tory weakness.9

D IAPHRAGMATIC PARALYSIS has been reported in a va-

riety of animal species including llamas, alpacas,
ponies, dogs, and cats.1–7 Proposed mechanisms include
pneumonia, trauma, neuropathy (degeneration of the
phrenic nerves that innervate the diaphragm muscles), my-
opathy or idiopathic causes.2 The paralysis may be uni-
lateral or bilateral, and may be temporary or permanent.
Unilateral diaphragmatic paralysis can be asymptomatic;
however, bilateral paralysis produces severe clinical signs
such as abdominal breathing or cyanosis with apnea.3–5
Diaphragmatic paralysis is more widely described in hu-
mans than in animals.8 Cervical or thoracic trauma and
pneumonia can affect the phrenic nerve passing through
the thoracic cavity. In addition, central or peripheral neu-
ropathies and myopathy can directly affect the diaphragm
From the Haemaru Referral Animal Hospital, Seongnam 463–050,
Republic of Korea (Choi, Lee, Kim, Keh, Lee, and Kim) and Department
of Veterinary Medical Imaging, College of Veterinary Medicine, Seoul
National University, Seoul 151–742, Republic of Korea (Choi).
Address correspondence and reprint requests to Mincheol Choi, at
the above address. E-mail: mcchoi@snu.ac.kr
Received March 14, 2013; accepted for publication June 25, 2013.
doi: 10.1111/vru.12126
Conventional diagnostic imaging methods for detect-
ing canine diaphragmatic paralysis include plain radiog-
raphy, fluoroscopy, and B-mode ultrasonography. In man,
thoracic radiographic diagnoses are based on minimal to
absent craniocaudal movement (<one-half of a vertebral
body) of the diaphragmatic crura during inspiration and
expiration.1, 6–7, 10, 11 However, the radiographic appearance
of the diaphragm in dogs varies, depending on patient po-
sitioning, x-ray beam centering, respiration, gravity, age,
and breed.6 Therefore a dynamic study is required to reach
a definitive diagnosis. Fluoroscopy is currently the refer-
ence standard method in dogs, but it has a disadvantage
in that both the patient and staff are exposed to radiation.
In addition, bilateral paralysis may be difficult to detect
with fluoroscopy if dogs have minimal or no diaphrag-
matic movement, compensatory abdominal contractions,
or if dogs have a paradoxical cranial displacement of the
flaccid diaphragm occurring at the end of inspiration.6 B-
mode ultrasonography has been reported to be as valu-
able as fluoroscopy in detecting diaphragmatic paralysis in
Vet Radiol Ultrasound, Vol. 55, No. 1, 2014, pp 102–108.

102
VOL. 55, NO. 1 DIAPHRAGMATIC PARALYSIS
humans, showing high sensitivity and specificity in the
assessment of abnormalities of diaphragmatic motion.12
However, it is sometimes difficult to identify the direction
of this motion, particularly in cases of tachypnea or with
decreased excursion.13
M-mode ultrasonography has previously been described
in humans for the evaluation of diaphragmatic movement
abnormalities, most notably in the pediatric population,13
and it has enabled generation of adequate information for
comparison with follow-up examinations. It can also be
used to quantitatively assess diaphragmatic movement by
using two parameters; direction of motion and amplitude
of excursion. Diaphragmatic movement is considered nor-
mal in humans if the diaphragm moves toward (i.e., an up-
ward M-mode tracing) the transducer during inspiration,
with excursion of greater than 4 mm and a difference in di-
aphragmatic excursion between both sides being less than
50%.14, 15 At the time of this study, no published reports
could be found describing the use of this technique in dogs.
The purpose of this study was to develop a method for
evaluating diaphragmatic movement using M-mode ultra-
sonography, and to describe findings in normal versus clin-
ically affected dogs.
Materials and Methods
Fifty-five clinically healthy adult dogs and two dogs with
diaphragmatic paralysis were recruited. Clinically, healthy
dogs were defined as those with no history or clinical signs
of respiratory disease. Thoracic radiography was performed
in all dogs. Dogs with abnormal radiographic findings were
excluded from the clinically healthy group. Dogs with sus-
pected diaphragmatic paralysis were also examined using
fluoroscopy. With owner consent, dogs included in the
study were then examined using B-mode and M-mode ul-
trasound with a 7.5 MHz convex transducer (Prosound
Alpha 6 and SSD-4000V, Hitachi Aloka Medical, Tokyo,
Japan).
For imaging the diaphragmatic crura, dogs were posi-
tioned in lateral recumbency and the hepatic parenchyma
was used as an acoustic window. When imaging from the
left side, the spleen, or empty stomach was also used as
an acoustic window. The probe was placed behind the last
rib at the most proximal part and directed subcostally and
cranially. The angle between the rib and probe was ad-
justed from 90 to 180 degrees until the ultrasound beam
was oriented perpendicular to the diaphragm crura in
the longitudinal plane. In addition, the probe was tilted
clockwise approximately 45 degrees as needed to permit
better visualization of both the diaphragmatic crura
(Fig. 1). Two veterinary radiologists (M. C., S. K.) indepen-
dently recorded movements of both diaphragmatic crura
using M-mode ultrasonography while checking the respi-
103
ratory phase. The M-mode cursor was positioned nearly
perpendicular to the visualized diaphragm and direction of
motion and amplitude of excursion (mm) were recorded.
For the measurement, the first caliper was placed at the
foot of the inspiration slope on the diaphragm echogenic
line and the second caliper was placed at the apex of this
slope. Several respiratory cycles were examined, and at least
three measurements were acquired for each value.
All statistical tests were selected and performed by one
of the authors (M. C). Data were expressed as mean ±
standard deviation (SD). Upper and lower limits of nor-
mal diaphragmatic excursion (i.e., ninety-fifth and fifth per-
centiles) as mean ± 1.96 × SD, and the right to left ratio
of diaphragmatic crus excursion values were calculated. In-
terobserver and intraobserver agreement was assessed with
kappa statistics with values of 1 indicating perfect agree-
ment. Statistical analyses were performed using a commer-
cially available software program (SPSS 18.0, SPSS Inc.,
Chicago) and a P < 0.05 was defined as significant.
Results
Normal Dogs
A total of 55 dogs were initially included in the normal
group based on normal clinical examination findings. Dogs
with thoracic radiographic diagnoses of pneumonia (n =
3), pulmonary edema (n = 3), metastasis (n = 2), and tra-
cheal collapse (n = 4) were excluded. The remaining 47
normal dogs were 1–17 years of age (median: 8.8 years),
and ranged from 2.2 to 15.3 kg in weight (median: 5 kg).
The diaphragm was consistently identified in B-mode im-
ages as an echogenic line between the interface of the lung
and liver. Both diaphragmatic motions were successfully as-
sessed using M-mode ultrasonography. Normal inspiratory
diaphragmatic movement was caudal or toward the probe,
with the corresponding M-mode trace being upward. The
expiratory trace was downward as the diaphragm moved
cranially or away from the probe (Fig. 2). The mean di-
aphragmatic excursion values of the right side were 7.01 ±
2.05 mm (n = 43, lower limit: 2.98 mm, upper limit: 11.03
mm) and those on the left side were 7.57 ± 2.41 mm (n
= 43, lower limit: 2.85 mm, upper limit: 12.30 mm). The
average of both diaphragmatic crura was 7.29 ± 2.24 mm
(n = 43, lower limit: 2.90 mm, upper limit: 11.68 mm). The
mean difference between the left and right diaphragmatic
excursion was 20% (n = 43, 0.5–47%). For interobserver
and intraobserver repeatability tests, all kappa values were
greater than 0.5 (P < 0.01).
Clinically Affected Dogs
Clinically affected dogs were a 13-year-old female York-
shire terrier dog (case 1) and a 3-year-old spayed female
104
FIG. 1. Schematic drawing showing the orientation of the ultrasound beam and diaphragmatic position for each respiratory phase. During inspiration
(solid line), the normal diaphragm moves caudally toward the transducer, while during expiration (dotted line), the diaphragm moves cranially away from the
transducer.
Maltese dog (case 2). Both presented with respiratory
distress. In case 1, dyspnea and paradoxical inward move-
ment of the left thorax were detected on physical examina-
tion. Thoracic radiographs were taken including right lat-
eral and ventrodorsal views at inspiration and expiration.
The left 8–11th ribs were fractured and there were also evi-
dence of a flail chest. Mildly increased opacity of the overall
lung lobes was considered to be as a result of hypoinflation.
An obvious small thoracic cavity due to cranial displace-
ment of the both diaphragmatic crura was detected on both
lateral and ventrodorsal views. No diaphragmatic move-
ment occurred between inspiratory and expiratory views
(Fig. 3). Contraction of the diaphragm was not evident dur-
ing fluoroscopy and paradoxical cranial displacement of the
flaccid diaphragm occurred during inspiration. A paradox-
ical movement of both diaphragmatic crura was also noted
upon M-mode ultrasonography (Fig. 4). The final diagnosis
was bilateral diaphragmatic paralysis of unknown origin.
In case 2, the patient showed tachypnea and had a history
of repeated episodes of pulmonary edema diagnosed at a
local hospital. On the referral day, thoracic radiographs
were taken including right lateral and ventrodorsal views
with inspiration and expiration. Unlike the history, there
was no abnormal finding in the pulmonary parenchyma.
On the lateral view, the left crus (which met the spine at
CHOI ET AL. 2014
T8) was displaced cranially relative to the right one (which
met the spine at T10) (Fig. 5) and showed less diaphrag-
matic movement of the left side for inspiration versus ex-
piration views (Right side: 1 intercostal space, Left: less
than 0.5 intercostal space). Unequal movement between
the crus was observed with fluoroscopy and the left crus
showed minimal movement during inspiration and expi-
ration. Upon M-mode ultrasonography, diaphragmatic ex-
cursion was 2.00 ± 0.42 mm on the left side and 4.05 ± 1.48
mm on the right side (Fig. 6) and the difference between the
left and right diaphragmatic excursion was 55%. The final
diagnosis was unilateral diaphragmatic paralysis on the left
side.
Conservative treatment was instituted in both cases. In
case 1, clinical signs and pulmonary infiltration were more
severely progressed on day 4. The patient went into res-
piratory arrest, suspected to be secondary to hypoxemia,
and subsequently died. In case 2, prednisolone (1 mg/kg,
twice a day) was maintained for 1 month and tapered for
5 months. Follow-up radiographic and fluoroscopic exami-
nation 1 month later confirmed caudal displacement of the
left diaphragmatic crus during inspiration. Upon M-mode
ultrasonography, diaphragmatic excursion was normalized
to a value of 6.07 ± 0.80 mm on the left side as well as on
the right side (Fig. 6C).
VOL. 55, NO. 1 DIAPHRAGMATIC PARALYSIS 105

FIG. 2. M-mode ultrasonography images illustrating normal versus paradoxical diaphragm movement in the longitudinal plane (A). During normal
inspiration, the M mode trace is upwards, as the diaphragm moves toward the probe. During expiration, the M mode trace is downwards as the diaphragm
moves away from the probe (black arrows). Paradoxical movement is defined as a reversed normal diaphragmatic motion (dotted arrows) (B).

Discussion radiographs were used to define normal lung status. Tho-

racic radiographic results do not necessarily indicate nor-
In the current study, the lower limit of diaphragmatic
mal pulmonary function. Also, a lower limit of diaphrag-
excursion was estimated to be 3 mm using M-mode ultra-
matic excursion (3 mm) occurred in normal dogs and this
sonography in normal small-sized dogs and a difference
might increase the number of false negative diagnoses in
in excursion between the both sides was less than 50%,
affected dogs. In fact, one normal dog measured near the
similar to humans. M-mode ultrasonography was easily
lower limit of 3 mm for both diaphragmatic excursions but
performed, repeatable within and between observers, and
the difference of the both sides was only 0.5%. Therefore,
measured diaphragmatic excursion values for clinically af-
applying both diaphragmatic excursion (mm) and differ-
fected dogs were lower than the lower limit in normal dogs.
ence (%) measurements would appear to be a useful di-
Techniques were also identified that aided in the evalua-
agnostic criterion for dogs with suspected diaphragmatic
tion of the diaphragm through ultrasonography. First, when
paralysis.
measuring diaphragmatic excursion, the ultrasound beam
Consistent with previously published studies, the prog-
should be as perpendicular as possible to the diaphragm.
nosis for diaphragmatic paralysis in the two affected dogs
Previous studies have found that deviations from this po-
of the current study was variable. In case 1, the dog had an
sitioning can result in a loss of accuracy and repeatability
inability to lower the diaphragm during active inspiration
of the diaphragmatic excursion measurements.16, 17 Second,
and most likely was unable to achieve a sufficient increase
there was little difference in the diaphragmatic excursion
in negative intra-thoracic pressure necessary to inflate the
between the diaphragmatic cupula and crus. Therefore, the
lungs. Multiple rib fractures induced a flail chest, and fur-
sonographer should take a dorsal approach with the ultra-
ther reduced the inspiratory volume, resulting in alveolar
sound beam as needed in order to visualize the diaphragm.
hypoventilation, and likely respiratory acidosis.5 Support-
The best diaphragmatic delineation in B-mode should be
ive therapy with oxygen supply and analgesic drugs im-
chosen before applying the M-mode measurements.18
proved the respiratory status somewhat, but the patient un-
There were some limitations to this study. Normal dogs
fortunately died on day 4. In case 2, conservative treatment
did not undergo pulmonary function tests. Instead, thoracic
106 CHOI ET AL. 2014

FIG. 3. Right lateral and ventrodorsal thoracic radiographs acquired during inspiration (A and C) and expiration (B and D) in Case 1. A small thoracic
cavity is seen on inspiration and there is bilateral cranial displacement of diaphragm. A soft tissue mass is visible on the left thoracic wall at the 4–6th intercostal
space. There is limited diaphragmatic movement identified for both sides.

FIG. 4. M-mode ultrasonography images for Case 1. Paradoxical movement was observed for both diaphragmatic crura (A: right, B: left). There was a
cranial diaphragmatic movement (away from the probe) upon inspiration due to bilateral diaphragmatic paralysis.

was effective, and recovery of diaphragmatic movement was
confirmed after one month. In both cases the cause of the
diaphragmatic paralysis was not determined.
In conclusion, findings from the current study indicated
that M-mode ultrasonography is a relatively simple and re-
peatable method for evaluating diaphragmatic movement
in dogs. This new method offers advantages for future clin-
ical application in that both qualitative and quantitative
information can be gained regarding diaphragmatic func-
tion. The method also allows measurement of response
VOL. 55, NO. 1 DIAPHRAGMATIC PARALYSIS 107

FIG. 5. Right lateral and ventrodorsal thoracic radiographs acquired during inspiration (A and C) and expiration (B and D) in Case 2. Cranial displacement
of the left diaphragmatic crus is seen. There is normal movement of the right side but minimal diaphragmatic movement of left side.

FIG. 6. M-mode ultrasonography images for Case 2. Decreased diaphragmatic motion of the left side was observed (B), compared to that of the right side
(A). Follow-up M-mode ultrasonography one month after treatment showed a recovery of diaphragmatic motion of the left side (C), and an increased similarity
in movement compared to the right side (A).

to treatment without the use of ionizing radiation. Future this promising technique in a larger number of clinically
studies are needed to assess the diagnostic sensitivity of affected dogs.

REFERENCES
1. Byers S, Barrington G, Nelson D, Haldorson G, Holt T, Callan R.
Neurological causes of diaphragmatic paralysis in 11 alpacas (Vicugna pacos).
J Vet Intern Med 2011;25:380–385.
2. Lori LL, Simpson AM, Han E. Pleural and extrapleural disease. In:
Ettinger SJ, Feldman EC (eds): Textbook of veterinary internal medicine,
7th ed. Philadelphia: WB Saunders, 2010;1125.
3. Bedenice D, Mazan MR, Kuehn H, Hoffman AM. Diaphragmatic
paralysis due to phrenic nerve degeneration in a llama. J Vet Intern Med
2002;16:603–606.
4. Amory H, Lomba F, Lekeux PM, Solal AN, Jauniaux TP, Desmecht
DJ. Bilateral diaphragmatic paralysis in a pony. J Am Vet Med Assoc
1994;205:587–591.
5. Greene CE, Basinger RR, Whitfield JB. Surgical management of bilat-
eral diaphragmatic paralysis in a dog. J Am Vet Med Assoc 1988;193:1542–
1544.
6. Park RD. The diaphragm. In: Thrall DE (ed): Textbook of vet-
erinary diagnostic radiology, 5th ed. Philadelphia: WB Saunders, 2007;
537.
108 CHOI ET AL.
7. Vignoli M, Toniato M, Rossi F, et al. Transient post-traumatic
hemidiaphragmatic paralysis in two cats. J Small Anim Pract 2002;43:312–
316.
8. Urvoas E, Pariente D, Fausser C, Lipsich J, Taleb R, Devictor D.
Diaphragmatic paralysis in children: diagnosis by TM-mode ultrasound.
Pediatr Radiol 1994;24:564–568.
9. Lloyd T, Tang YM, Benson MD, King S. Diaphragmatic paralysis: the
use of M mode ultrasound for diagnosis in adults. Spinal Cord 2006;44:505–
508.
10. Epelman M, Navarro OM, Daneman A, Miller SF. M-mode
sonography of diaphragmatic motion: description of technique and
experience in 278 pediatric patients. Pediatr Radiol 2005;35:661–
667.
11. Boussuges A, Gole Y, Blanc P. Diaphragmatic motion studied by m-
mode ultrasonography: methods, reproducibility, and normal values. Chest
2009;135:391–400.
12. Gibson GJ. Diaphragmatic paresis: Pathophysiology, clinical features,
and investigation. Thorax 1989;44:960–970.
2014
13. Chavhan GB, Babyn PS, Cohen RA, Langer JC. Multimodality imag-
ing of the pediatric diaphragm: anatomy and pathologic conditions. Radio-
graphics 2010;30:1797–1817.
14. Houston JG, Morris AD, Howie CA, Reid JL, McMillan N. Technical
report: quantitative assessment of diaphragmatic movement–a reproducible
method using ultrasound. Clin Radiol 1992;46:405–407.
15. Balaji S, Kunovsky P, Sullivan I. Ultrasound in the diagnosis of di-
aphragmatic paralysis after operation for congenital heart disease. Br Heart
J 1990;64:20–22.
16. Harris RS, Giovannetti M, Kim BK. Normal ventilatory movement
of the right hemidiaphragm studied by ultrasonography and pneumotachog-
raphy. Radiology 1983;146:141–144.
17. Matamis D, Soilemezi E, Tsagourias M, et al. Sonographic evaluation
of the diaphragm in critically ill patients. Technique and clinical applications.
Intensive Care Med 2013;39:801–810.
18. Young PN, Gorgacz EJ, Barsanti JA. Respiratory failure associated
with diaphragmatic paralysis in a cat. J Am Anim Hosp Assoc 1980;16:933–
936.