AJR:175, September 2000

727

Diagnosis of Primary Versus Secondary Achalasia:

Reassessment of Clinical and Radiographic Criteria

OBJECTIVE.

Our purpose was to reassess the usefulness of barium studies and variousclinical parameters for differentiating primary from secondary achalasia.

MATERIALS AND METHODS.

Radiology files from 1989 through 1999 revealed 29patients with primary achalasia and 10 with secondary achalasia (caused by carcinoma of theesophagus in three, of the gastric cardia in three, of the lung in three, and of the uterus in one)who met our study criteria. The radiographs were reviewed to determine the morphologic fea-tures of the narrowed distal esophageal segment and gastric cardia and fundus. Medicalrecords were also reviewed to determine the clinical presentation; endoscopic, manometric,and surgical findings; and treatment.

RESULTS.

The mean patient age was 53 years in primary achalasia versus 69 years insecondary achalasia (

p

= 0.03). The mean duration of dysphagia was 4.5 years in primaryachalasia versus 1.9 months in secondary achalasia (

p

< 0.0001). The narrowed distal esoph-ageal segment had a mean length of 1.9 cm in primary achalasia versus 4.4 cm in secondaryachalasia (

p

< 0.0001), and the esophagus had a mean diameter of 6.2 cm in primary achala-sia versus 4.1 cm in secondary achalasia (

p

< 0.0001). The narrowed segment was eccentricor nodular or had abrupt proximal borders in only four of 10 patients with secondary achala-sia, and evidence of tumor was present in the gastric fundus in only three.

CONCLUSION.

When findings of achalasia are present on barium studies, a narroweddistal esophageal segment longer than 3.5 cm with little or no proximal dilatation in a patientwith recent onset of dysphagia should be considered highly suggestive of secondary achala-sia, even in the absence of other suspicious radiographic findings.

chalasia is a well-known esoph-ageal motility disorder character-ized by absent primary peristalsis

and incomplete relaxation of the lower esoph-ageal sphincter [1]. Most patients have primary(idiopathic) achalasia caused by loss of theganglion cells in the esophageal myentericplexuses [2, 3]. However, others have second-ary achalasia (pseudoachalasia) caused by ma-lignant tumor at the gastroesophageal junction[4–9] or, less commonly, by benign conditionssuch as Chagas’ disease [10]. Nearly 75% ofpatients with secondary achalasia are found tohave underlying carcinoma of the cardia [6],but secondary achalasia may also be caused bycarcinoma of the esophagus or by other malig-nant tumors that metastasize to the mediasti-num or gastroesophageal junction, includingcarcinoma of the lung, breast, pancreas, uterus,and prostate gland [4, 7–9].

Primary achalasia is characterized on bar-ium studies by absent primary peristalsis and

smooth, tapered narrowing of the distalesophagus caused by incomplete relaxationof the lower esophageal sphincter [11]. How-ever, in secondary achalasia, barium studiesmay also reveal eccentricity, nodularity, an-gulation, straightening, or proximal shoul-dering of the narrowed segment [4, 7, 8, 12,13]. In one report, it was suggested that thenarrowed segment may be longer in second-ary than in primary achalasia [12]. Second-ary achalasia should also be suspected ifbarium studies reveal tumor at the gastriccardia [4, 12, 13].

Nevertheless, little data are availableabout the usefulness of barium studies in dif-ferentiating primary from secondary achala-sia. In the two largest series in the literature,it was possible to distinguish these condi-tions on barium studies in only six (46%) of13 patients [6, 14]. We therefore performed aretrospective investigation of patients withprimary and secondary achalasia to reassess

Courtney A. Woodfield

1

Marc S. LevineStephen E. RubesinCurtis P. LanglotzIgor Laufer

Received January 14, 2000; accepted after revision February 16, 2000.

1

All authors: Department of Radiology, Hospital of the University of Pennsylvania, 3400 Spruce St., Philadelphia, PA 19104. Address correspondence to M. S. Levine.

AJR

2000;175:727–731

0361–803X/00/1753–727

© American Roentgen Ray Society

A

728

AJR:175, September 2000

Woodfield et al.

the usefulness of barium studies and variousclinical parameters for differentiating theseconditions.

Materials and Methods

Computerized radiology files at our universityhospital from 1989 through 1999 and radiologylogs at our affiliated Veterans Affairs medical cen-ter from 1995 through 1999 revealed 150 patientswith a diagnosis of achalasia on barium studies.Seventy-two of these 150 patients were excludedfrom our analysis because of known treatment forachalasia (e.g., pneumatic dilatation, botulinumtoxin injection, or surgical myotomy) before un-dergoing any barium studies at our hospital, and39 were excluded because medical records wereunavailable (36 patients) or clinical follow-up wasinadequate to establish the diagnosis (three pa-tients). The remaining 39 patients constituted ourstudy group.

On the basis of the endoscopic, manometric,CT, and surgical findings, 29 patients (74%) had afinal diagnosis of primary achalasia, and 10 (26%)had a final diagnosis of secondary achalasiacaused by carcinoma of the esophagus in three pa-tients, carcinoma of the gastric cardia in three, andmetastases to the mediastinum or gastroesoph-ageal junction from carcinoma of the lung in threeand from carcinoma of the uterus in one.

All 39 patients underwent barium studies, in-cluding double-contrast esophagography in 11,single-contrast esophagography in five, double-contrast upper gastrointestinal examinations in 17,and single-contrast upper gastrointestinal exami-nations in six. In all 39 patients, the radiographicreports described absent primary peristalsis in theesophagus on fluoroscopy and a segment of distalesophageal narrowing that extended to the gastro-esophageal junction. The correct diagnosis wassuggested on the original radiology reports in all10 patients with secondary achalasia.

The radiographs from these 39 studies were re-viewed in a blinded fashion to determine the de-gree of esophageal dilatation at its widest pointand to evaluate the morphologic features of thenarrowed distal esophageal segment, includingsymmetry (symmetric versus eccentric), contour(smooth versus nodular or ulcerated), proximalborders (tapered versus abrupt or shouldered), andlength (measured from the proximal border of thenarrowed segment to the gastroesophageal junc-tion, not accounting for radiographic magnifica-tion). When sufficient barium entered the stomach,the gastric cardia and fundus were also evaluatedfor evidence of tumor in this region.

Medical, radiologic, and endoscopic recordswere also reviewed to determine the clinical pre-sentation as well as the endoscopic, manometric,CT, and surgical findings.

Univariate statistical analysis was performed onall major study variables. Wilcoxon’s rank sumtest was performed using JMP statistical analysissoftware (SAS Institute, Cary, NC) to determine

whether the patient’s age, the duration of dyspha-gia, the length of the narrowed distal esophagealsegment, or the diameter of the proximal esopha-gus was significantly associated with achalasia eti-ology (i.e., primary versus secondary achalasia).

Results

Clinical Findings

Primary achalasia.

—Sixteen of the 29 pa-tients with primary achalasia were womenand 13 were men. The mean age was 53years (range, 22–87 years); 11 patients(38%) were more than 60 years old. All 29patients presented with dysphagia, whichhad a mean duration of 4.5 years (range, 0.1–20 years); 28 patients (97%) had dysphagiafor 1 year or longer. Five patients had weightloss, with a mean loss of 8.2 kg (range, 3.6–16 kg) over a mean period of 13 months(range, 2–36 months).

Secondary achalasi

a.—Nine of the 10 pa-tients with secondary achalasia were menand one was a woman. The mean age was 69years (range, 48–87 years); eight patients(80%) were more than 60 years old. All 10patients presented with dysphagia, whichhad a mean duration of 1.9 months (range,0.5–4 months). Patients with secondaryachalasia were significantly more likely to beolder (

p

= 0.03) and to have a shorter dura-tion of dysphagia (

p

< 0.0001) than patientswith primary achalasia (Table 1). Seven pa-tients had weight loss, with a mean loss of10.5 kg (range, 2.7–30 kg) over a mean pe-riod of 5 months (range, 0.5–12 months).

Radiographic Findings

Primary achalasia

.—In all 29 patientswith primary achalasia, barium studies re-vealed smooth symmetric, tapered narrowingof the distal esophagus that extended to thegastroesophageal junction (Figs. 1 and 2).The narrowed segment had a mean length of1.9 cm (range, 0.7–3.5 cm). The esophagusabove the narrowed segment had a mean di-ameter of 6.2 cm (range, 4–10 cm) and wasgreater than 4 cm in diameter in 26 patients

(90%). In two patients, the distal esophagushad a tortuous (i.e., sigmoid) configuration.The gastric cardia and fundus appeared nor-mal in 10 patients (34%) but could not be ad-equately evaluated because of delayedemptying of barium from the esophagus inthe remaining 19 patients (66%).

Secondary achalasia

.—In six (60%) of 10patients with secondary achalasia, barium stud-ies revealed smooth symmetric, tapered nar-rowing of the distal esophagus (Figs. 3 and 4).The remaining four patients (40%) had eccen-tric narrowing of the distal esophagus (Fig.5A), with abrupt proximal borders in one, nod-ularity in one, and straightening in one. Thenarrowed segment had a mean length of 4.4 cm(range, 2.5–5.0 cm) and was longer than 3.5 cmin eight patients (80%) (Figs. 3–5). The esoph-agus above the narrowed segment had a meandiameter of 4.1 cm (range, 3.5–6 cm) and was4 cm or less in diameter in eight patients(80%). Patients with secondary achalasia weresignificantly more likely to have a longer seg-ment of narrowing (

p

< 0.0001) and to have aless dilated proximal esophagus (

p

< 0.0001)than patients with primary achalasia (Table 1).

One patient also had an annular lesionwith abrupt shelflike borders in the upperesophagus caused by esophageal carcinoma(Fig. 5B). Secondary achalasia in this patientpresumably resulted from the spread of tu-mor via lymphatics in the esophageal wall tothe gastroesophageal junction.

The gastric cardia and fundus appearedabnormal in three patients (30%) with sec-ondary achalasia. Two had carcinoma of thecardia; barium studies revealed lobulatedfundal folds in one and encasement of thefundus by tumor in the other. In one patientwith esophageal carcinoma, a barium studyrevealed nodularity of the gastric fundus. Intwo other patients, barium studies revealed anormal-appearing cardia and fundus. In theremaining five patients (including one withcarcinoma of the cardia), the cardia and fun-dus could not be adequately evaluated be-cause of delayed emptying of barium fromthe esophagus.

TABLE 1 Major Variables of Primary and Secondary Achalasia in 30 Patients

VariablePrimary Achalasia

(20 Patients)Secondary Achalasia

(10 Patients)p

Age (years) 53 ± 19 69 ± 12 0.03Duration of dysphagia (months) 54 ± 52 1.9 ± 1.2 < 0.0001Length of narrowing (cm) 1.9 ± 0.78 4.4 ± 0.88 < 0.0001Diameter of proximal esophagus (cm) 6.2 ± 1.5 4.1 ± 0.76 < 0.0001

Radiography of Primary Versus Secondary Achalasia

AJR:175, September 2000

729

Endoscopic, Manometric, CT, and Surgical Findings

Primary achalasia

.—Twenty-five of the 29patients with primary achalasia had typicalfindings of achalasia on manometry [1, 15]. Inall 29 patients, endoscopy revealed a closedlower esophageal sphincter that opened in re-sponse to the advancing endoscope, allowing itto pass into the gastric fundus [15].

Secondary achalasia

.—Eight of the 10 pa-tients with secondary achalasia underwent en-doscopy, which revealed a closed loweresophageal sphincter in all cases; the endoscopecould not be advanced into the stomach in fourof these patients, a finding that has been associ-ated with secondary achalasia [6, 15–17]. Threepatients had esophageal carcinoma at endos-copy, with infiltrative lesions in the distal esoph-agus in two and in the upper esophagus in one;endoscopic biopsy specimens revealed squa-mous cell carcinoma in all three patients. Threeother patients had carcinoma of the cardia onendoscopy, with polypoid masses in the gastric

fundus; endoscopic biopsy specimens revealedcarcinoma of the cardia in two of these patients.The third had carcinoma of the cardia at sur-gery. In three patients with lung carcinoma,chest CT scans revealed mediastinal adenopa-thy and mediastinal invasion by tumor. In theremaining patient with endometrial carcinoma,an abdominal CT scan revealed widespread in-traperitoneal metastases, and a bone scan re-vealed diffuse osseous metastases. Althoughthis patient did not have a chest CT scan, shewas presumed to have secondary achalasia be-cause of her widely disseminated endometrialcarcinoma, advanced age (87 years), and shortduration of dysphagia (3 months).

Discussion

In our study, barium studies revealed classicfindings of secondary achalasia with an eccen-tric, nodular, or shouldered segment of distalesophageal narrowing (Fig. 5A) in only 40% of

patients with this condition. In the remaining60%, the narrowed segment was smooth andsymmetric with tapered proximal borders(Figs. 3 and 4). Therefore, secondary achalasiawould not be suspected in most cases solely onthe basis of classic radiologic criteria. However,the narrowed distal esophageal segment waslonger than 3.5 cm in 80% of patients with sec-ondary achalasia, and an unusually long seg-ment of narrowing was the only suspiciousfinding in 40% (Figs. 3 and 4). In contrast, thenarrowed segment was 3.5 cm or shorter in allpatients with primary achalasia (Figs. 1 and 2).Therefore, the length of the narrowed distalesophageal segment was a useful and statisti-cally significant criterion for differentiatingsecondary achalasia from primary achalasia onbarium studies (

p

< 0.0001).In our series, the degree of esophageal dila-

tation above the narrowed segment was also astatistically significant criterion for differentiat-ing secondary achalasia from primary achalasia

Fig. 1.—50-year-old man with primaryachalasia. Spot radiograph from double-contrast barium study shows 1-cm-longsmooth, tapered narrowing (straightarrow ) of distal esophagus with esoph-ageal diameter proximally of 6 cm. Notestanding column of barium (curved arrow )on this upright view. Short length of nar-rowed segment is characteristic of primaryachalasia.

Fig. 2.—23-year-old woman with primary achalasia. Spot ra-diograph from double-contrast barium study shows 3.5-cm-long, gradually tapered segment of narrowing (straightarrows) in distal esophagus with esophageal diameter proxi-mally of 7 cm and standing column of barium (curved arrow ).This was longest segment of narrowing shown on radiogra-phy in a patient with primary achalasia.

Fig. 3.—60-year-old man with secondaryachalasia caused by lung carcinoma. Spot ra-diograph from single-contrast barium studyshows 5-cm-long symmetric, tapered narrow-ing (arrows) of distal esophagus with esoph-ageal diameter proximally of 6 cm. Note fineirregularity of contour of distal esophagusabove narrowed segment caused by super-imposed infection with Candida esophagitisorganisms proven on endoscopy.

730

AJR:175, September 2000

Woodfield et al.

(

p

< 0.0001). The diameter of the esophagus atits widest point was 4 cm or less in 80% of pa-tients with secondary achalasia, whereas the di-ameter of the esophagus was greater than 4 cmin 90% of patients with primary achalasia. Thegreater degree of esophageal dilatation in pa-tients with primary achalasia was presumablyrelated to the more gradual course of the dis-ease that allowed the esophagus to progres-sively dilate over a period of years. In fact, bothpatients who had a tortuous (i.e., sigmoid) dis-tal esophagus were found to have primaryachalasia with relatively long-standing disease.

A limitation of our study is the variable ef-fect of magnification on our radiographicmeasurements of the narrowed distal esoph-ageal segment or dilated proximal esophagusin patients with primary or secondary achala-sia, depending on the height of the fluoro-scopic tower above the examining table. Thisvariable could create a potential bias ifgreater magnification occurred primarily inone group or the other. However, the degreeof magnification was in no way related to pa-tient selection, so this variable should nothave had a significant effect on our findings.

When findings of achalasia are present onbarium studies, it is important to evaluate thegastric cardia and fundus to rule out an under-lying malignant tumor at the gastroesophagealjunction as the cause of these findings [4, 6, 13,

Fig. 4.—87-year-old woman withsecondary achalasia caused by car-cinoma of uterus. Spot radiographfrom double-contrast barium studyshows 4-cm-long smooth, taperednarrowing (arrows) of distal esoph-agus with esophageal diameterproximally of 3.5 cm. As in Figure 3, anarrowed segment longer than 3.5cm should be considered highly sug-gestive of secondary achalasia,even lacking other suspicious radio-graphic findings.

Fig. 5.—63-year-old man with sec-ondary achalasia caused by carci-noma of esophagus.A, Spot radiograph from double-con-trast barium study shows 4-cm-longeccentric, tapered narrowing (arrows)of distal esophagus with esophagealdiameter proximally of 4 cm.B, Additional spot radiograph showsannular carcinoma with relativelyabrupt, shelflike margins (arrows) inupper thoracic esophagus.

BA

Radiography of Primary Versus Secondary Achalasia

AJR:175, September 2000

731

18]. In our series, however, the cardia and fun-dus could not be adequately evaluated radio-graphically in 66% of patients with primaryachalasia and in 50% with secondary achalasiabecause of delayed emptying of barium fromthe esophagus. Therefore, it is important to beaware of the limitations of barium studies inevaluating the cardia and fundus in patientswith suspected achalasia.

In the past, some investigators have advo-cated amyl nitrite inhalation as a simple testfor differentiating primary and secondaryachalasia on barium studies. It has beenshown that inhalation of amyl nitrite, asmooth-muscle relaxant, has no effect on thenarrowed distal esophageal segment in sec-ondary achalasia but causes a measurable in-crease of 2 mm or more in the caliber of thissegment in primary achalasia [19]. Neverthe-less, this technique has not gained wide-spread acceptance.

Although our investigation focused on theusefulness of barium studies for differentiat-ing the two forms of achalasia, CT may alsobe useful in these patients. CT typically re-veals little or no esophageal wall thickeningand no evidence of a mass at the cardia in pa-tients with primary achalasia [20–22]. Insome cases, however, CT may reveal apseudomass at the cardia in patients withouttumor because of inadequate distention ofthis region [23]. In contrast, CT may showasymmetric thickening of the distal esoph-ageal wall, a soft-tissue mass at the cardia, ormediastinal adenopathy in patients with sec-ondary achalasia [21]. CT may also be help-ful for identifying the site of the primarytumor in patients with secondary achalasiacaused by remote tumors.

Various clinical parameters are also pur-ported to be useful for differentiating pri-mary achalasia from secondary achalasia,including the age of the patient, the durationof dysphagia, and substantial weight loss.Primary achalasia is more likely to occur inyounger patients (<50 years old) with long-standing dysphagia (>1 year) and little or noweight loss (<7 kg) [15, 18], whereas sec-ondary achalasia is more likely to occur inolder patients (>60 years old) with recent on-set of dysphagia (<6 months) and substantialweight loss (>7 kg) [14]. Nevertheless, over-

lap in the clinical presentation has been re-ported for all these parameters [17, 24]. Inour series, the duration of dysphagia was astatistically significant clinical criterion fordifferentiating secondary achalasia from pri-mary achalasia (

p

< 0.0001); all patientswith secondary achalasia had dysphagia for4 months or less, whereas 97% of patientswith primary achalasia had dysphagia for 1year or more. The age of the patient was alsoa statistically significant but somewhat lessuseful criterion for differentiating these con-ditions (

p

= 0.03); 80% of patients with sec-ondary achalasia and 38% with primaryachalasia were more than 60 years old. Intwo previously published series, 28–30% ofpatients with primary achalasia were alsofound to be more than 60 years old [17, 24],limiting the usefulness of this criterion.

In conclusion, only 40% of patients in our se-ries had classic radiographic features of second-ary achalasia such as eccentricity, nodularity, orshouldering of the narrowed distal esophagealsegment, or suspicious findings in the region ofthe gastric cardia or fundus. Instead, the mostuseful criteria for differentiating secondaryfrom primary achalasia were the length of thenarrowed segment and the degree of proximaldilatation, and the most useful clinical criterionwas the duration of dysphagia. When findingsof achalasia are present on barium studies, anarrowed distal esophageal segment longer than3.5 cm with little or no proximal dilatation in apatient with recent onset of dysphagia should beconsidered highly suggestive of secondaryachalasia, even in the absence of other suspi-cious radiographic findings.

References

1. Katz PO, Castell DO. Review: esophageal motilitydisorders.

Am J Med Sci

1985

;290:61–69 2. Cassella RR, Brown AL, Sayre GP, Ellis FH.

Achalasia of the esophagus: pathologic and etio-logic considerations.

Ann Surg

1964

;160:474–4873. Csendes A, Smok G, Braghetto I, Ramirez C, Ve-

lasco N, Henriquez A. Gastroesophageal sphincterpressure and histological changes in distal esopha-gus in patients with achalasia of the esophagus.

Dig Dis Sci

1985

;30:941–9454. Lawson TL, Dodds WJ. Infiltrating carcinoma simu-

lating achalasia.

Gastrointest Radiol

1976

;1:245–248 5. McCallum RW. Esophageal achalasia secondary to

gastric carcinoma: report of a case and review of

the literature.

Am J Gastroenterol

1979

;71:24–29 6. Kahrilas PJ, Kishk SM, Helm JF, Dodds WJ, Harig

JM, Hogan WJ. Comparison of pseudoachalasiaand achalasia.

Am J Med

1987

;82:439–4467. Feczko PJ, Halpert RD. Achalasia secondary to

nongastrointestinal malignancies.

Gastrointest Ra-diol

1985

;10:273–276 8. Joffe N. Right-angled narrowing of the distal

oesophagus secondary to carcinoma of the tail ofthe pancreas.

Clin Radiol

1979

;30:33–37 9. Eaves R, Lambert J, Rees J, King RWF. Achalasia

secondary to carcinoma of the prostate.

Dig Dis Sci

1983

;28:278–28410. Ferreira-Santos R. Aperistalsis of the esophagus

and colon etiologically related to Chagas’ disease.

Am J Dig Dis

1961

;6:700–72611. Ott DJ. Motility disorders. In: Gore RM, Levine

MS, Laufer I, eds.

Textbook of gastrointestinal ra-diology.

Philadelphia: Saunders,

1994

:346–35912. Seaman WB, Wells J, Flood CA. Diagnostic prob-

lems of esophageal cancer: relationship to achala-sia and hiatus hernia.

AJR

1963

;90:778–791 13. Marshak RH, Eliasoph J. Cardiospasm or carci-

noma? The roentgen findings.

Am J Dig Dis

1957

;2:11–25

14. Tucker HJ, Snape WJ, Cohen S. Achalasia second-ary to carcinoma: manometric and clinical fea-tures.

Ann Intern Med

1978

;89:315–318 15. Reynolds JC, Parkman HP. Achalasia.

Gastroen-terol Clin North Am

1989

;18:223–25516. Rozman RW, Achkar E. Features distinguishing

secondary achalasia from primary achalasia.

Am JGastroenterol

1990

;85:1327–133017. Tracey JP, Traube M. Difficulties in the diagnosis

of pseudoachalasia.

Am J Gastroenterol

1994

;89:2014–2018

18. Levine MS. Other malignant tumors. In: Gore RM,Levine MS, Laufer I, eds.

Textbook of gastrointesti-nal radiology

. Philadelphia: Saunders,

1994

:479–49819. Dodds WJ, Stewart ET, Kishk SM, Kahrilas PJ,

Hogan WJ. Radiologic amyl nitrite test for distin-guishing pseudoachalasia from idiopathic achala-sia.

AJR

1986

;146:21–2320. Rabushka LS, Fishman EK, Kuhlman JE. CT eval-

uation of achalasia.

J Comput Assist Tomogr

1991

;15:434–439

21. Carter M, Deckmann RC, Smith RC, Burrell MI,Traube M. Differentiation of achalasia frompseudoachalasia by computed tomography.

Am JGastroenterol

1997

;92:624–62822. Tishler JM, Shin MS, Stanley RJ, Koehler RE. CT

of the thorax in patients with achalasia.

Dig Dis Sci

1983

;28:692–69723. Marks WM, Callen PW, Moss AA. Gastroesoph-

ageal region: source of confusion on CT.

AJR

1981

;136:359–362

24. Sandler RS, Bozymski EM, Orlando RC. Failureof clinical criteria to distinguish between primaryachalasia and achalasia secondary to tumor.

DigDis Sci

1982

;27:209–213