The Dysphagia Short Questionnaire

Transcription

1 SPINE Volume 37, Number 11, pp , Lippincott Williams & Wilkins HEALTH SERVICES RESEARCH The Dysphagia Short Questionnaire An Instrument for Evaluation of Dysphagia: A Validation Study With 12 Months Follow-up After Anterior Cervical Spine Surgery Martin Skeppholm, MD, * Catarina Ingebro, SLP, Therese Engström, SLP, and Claes Olerud, MD, PhD Study Design. Prospective clinical validation study of questionnaire to assess dysphagia. Objective. To test validity and reliability of Dysphagia Short Questionnaire (DSQ), and also to determine levels of dysphagia over time after anterior cervical spine surgery (ACSS). Summary of Background Data. Dysphagia is common after ACSS but reports on the incidence vary widely between 1% and 79%, indicating an evaluation problem. Several tools for evaluation of dysphagia exist but common features are that they are cumbersome to use and usually are designed for patients with neurological or malignant diseases in the neck region. Others are not validated, for example, the Bazaz score. There is, thus, a need for a more adapted tool to evaluate dysphagia in patients undergoing ACSS. Methods. The DSQ was constructed in collaboration with a group of ear-nose-and-throat specialists. In a first validation study, 45 patients with stationary dysphagia for various reasons completed the DSQ twice 2 weeks apart, the M.D. Anderson Dysphagia Inventory (MDADI), the Bazaz score, and a quality-of-life score, the EQ-5D. To evaluate the utility of the DSQ, a second validation study was performed, where 111 subjects undergoing ACSS for degenerative disk disease completed the form preoperatively and at 4 weeks, 3 months, and 1 year after surgery. Results. In the first study, the DSQ correlated to MDADI ( r = 0.59) and showed good reproducibility. The Bazaz score did not correlate to the DSQ, the MDADI, or the EQ-5 D. In the second study, dysphagia was present in a few patients already preoperatively. At From the * Karolinska Institutet, Stockholm Spine Center, Löwenströmska Hospital, Upplands Väsby, Stockholm, Sweden ; Karolinska Institutet, CLINTEC, Division of Speech and Language Pathology; and Department of Orthopedics, Uppsala University Hospital, Uppsala, Sweden. Acknowledgment date: May 23, First revision date: August 17, Second revision date: September 14, Acceptance date: September 19, The study was approved by the Regional Ethics Committee in Stockholm: 2006/ /4 and 2007/336-31/3. The legal regulatory status of the device(s)/drug(s) that is/are the subject of this manuscript is not applicable in my country. Corporate/Industry funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript. Address correspondence and reprint request to Martin Skeppholm, MD, Martin Skeppholm, Stockholm Spine Center, Löwenströmska Sjukhuset, SE Upplands Väsby, Sweden; Martin.Skeppholm@spinecenter.se 4 weeks, 85% of the patients reported dysphagia. The level had dropped significantly at 3 months and had returned to baseline levels at 1 year. Conclusion. We consider the DSQ to be a validated tool for the assessment of dysphagia in ACSS patients. Dysphagia after ACSS for cervical spondylosis is common but the symptoms on a group level are not very severe and are also temporary. Key words: spine surgery, dysphagia, validation, questionnaire. Spine 2012 ; 37 : A common complication to anterior cervical spine surgery (ACSS) is dysphagia. The mechanism is not fully understood, but reasonable explanations could be preoperative pressure on soft tissue structures, bulky implants, and damage to local innervation. 1, 2 Formation of scar tissue in the area of surgery postoperatively may also play a role. 3 Reports on its incidence vary widely between 1% and 79%. 1, 4 12 One reason for the wide variation may be different definitions and measurements of dysphagia in the various publications. Another explanation is different time intervals of evaluation after surgery because the presence of dysphagia decreases with time in this population. However, some studies have shown persistent dysphagia several years after surgery. 13, 14 Furthermore, there is a tendency toward higher incidence of dysphagia when patient self-report and in prospective studies whereas objective methods of measurement, for example, video fluoroscopy or manometry, may underestimate the problem. 8 To evaluate dysphagia from a patient-oriented perspective, there are some validated patient self-report questionnaires, for example, the M.D. Anderson Dysphagia Inventory (MDADI) 15 and SWAL-QOL. 16 These are relatively complex instruments and aim mostly at patients with neurological or malignant diseases in the neck region with significant swallowing problems. It can be assumed that ACSS patients experience less severe symptoms of dysphagia and therefore constitute a different subgroup. Bazaz et al 10 presented in an article on dysphagia after ACSS a much-simplified score, which later has been used in several other publications on similar patients. However, the Bazaz score has, to our knowledge, never been validated. The aim of this study was to design and validate an easy-to-use patient self-report dysphagia questionnaire and to apply it to a group of patients undergoing anterior cervical disc surgery. DOI: /BRS.0b013e31823a7a5b May 2012

2 DESIGN OF THE QUESTIONNAIRE The Dysphagia Short Questionnaire (DSQ) was designed in collaboration with 4 ear-nose-and-throat specialists at the Department of Ear-Nose-and-Throat diseases, Uppsala University Hospital, with vast experience of treating patients with dysphagia. A number of common symptoms that could be expected in patients with oropharyngeal dysphagia were listed. After discussions within the group, the symptom list was reduced to finally contain five key questions, each covering different aspects of dysphagia, with various response categories representing different grades of clinical severity. Each response category was given a weight, that is, the number of points, in accordance with how serious and important this particular problem was considered to be for the patient. The questionnaire was again discussed with the group of ear-nose-and-throat physicians to get their opinion on relevance of the items, statements, and relative weights, and adjustments were made according to suggestions. A score was calculated by summing up the points given for each item to a maximum of 18 points, where lower scores represent milder symptoms and vice versa (see Appendix, Supplemental Digital Content 1, available at: The questionnaire was tested on 10 patients who underwent ACSS to ensure that even patients with clinically mild symptoms of dysphagia were detected. We also wanted to know that no misunderstandings would appear among the tested patients, and also the approximate time for filling out the questionnaire, which typically took less than 1 minute. In the validation process, we wanted to address the following questions: Does the DSQ measure dysphagia? Does higher score represent more severe dysphagia and lower score milder? Will repeated applications result in similar scores? Is there a correlation between the DSQ and general health expressed as quality of life (QOL)? Is the DSQ usable in a cohort of patients undergoing ACSS? The MDADI was chosen for comparison as this is an already-validated scoring system, whereas the Bazaz score was included as this has previously been applied in several studies on cervical spine surgery. Both the applied scores were translated into Swedish and retranslated into English by 2 authorized translators. The study was approved by the regional ethics committee in Stockholm. PATIENTS AND METHODS The First Validation Study Forty-five patients, including 22 women, undergoing treatment of dysphagia for various reasons were included after informed consent. The dysphagia should be stationary: no major change in the underlying disease could be expected during the 2 to 3 weeks the validation process lasted. Their average age was 64.8 years (SD = 10.4). Their various diagnoses are listed in Table 1. The test patients were recruited with the help of 2 speech-language pathologists (SLP) working at the unit for dysphagia, Department of Speech Pathology, Karolinska University Hospital, Solna. The SLPs also instructed the patients how to fill out the various instruments. The DSQ was TABLE 1. The Various Diagnoses of the Patients in the First Validation Study Group Diagnosis compared with the MDADI and the Bazaz scores and correlations between them were computed statistically. The DSQ was also submitted to a test-retest situation to test its reproducibility. For health-related QOL, the EQ-5D was used All patients were examined by the 2 SLPs. At the first visit (T1), the patients completed our DSQ, the MDADI, and the Bazaz score. At the second visit (T2), a couple of weeks later the patients once again completed the DSQ and a healthrelated QOL score: EQ-5D. Five patients did not participate at T2; therefore, only 40 patients are included in the testretest analysis and the correlation to health-related QOL. The Second Validation Study One hundred eleven patients undergoing surgery for degenerative disk disease were included after informed consent. Inclusion criteria were as follows: at least 3 months of radicular pain not responding to conservative therapy and corresponding findings on magnetic resonance imaging at 1 or 2 levels, aged 25 to 60 years. Exclusion criteria were congenital malformation, severe degeneration, previous surgery for degenerative disk disease, inflammatory spine disease, and drug or alcohol abuse. Average age was 46.9 years (SD = 6.7), 56 (50%) were women, and 35 (32%) were smokers. They were evaluated with DSQ preoperatively and at 4 weeks, 3 months, and 1 year after surgery. Statistics All analyses were done using the Statistica 9.0 software (StatSoft Inc., Tulsa, OK). For the validation test of DSQ, Spearman rank test and nonparametric Gamma test for correlations were used. P < 0.05 was considered as statistically significant. Cronbach alpha coefficient was computed for the reliability test and a Bland-Altman diagram was constructed with a 95% confidence interval. Cronbach alpha coefficient was considered good if the value was higher than 0.6 and very good if it was more than 0.8. Wilcoxon signed-rank test for ordinal data was used for the second validation study. RESULTS No. of Patients Malignancy of the neck region 21 Neurological disease affecting the neck region 14 Miscellaneous 4 Xerostomia 2 Unknown 4 The First Validation Study DSQ ranged from 2 to 13 with an average of 6.3 (SD = 2.7) and was fairly normally distributed ( Figure 1 ) as was the MDADI ( Figure 2 ), while the distribution of the Bazaz score Spine 997

3 was more symmetrical with 17 of the patients classified as mild, 14 as moderate, and 14 as severe. There was a significant correlation ( r = 0.59, P < 0.01) between the DSQ and the MDADI as is seen in Figure 3. There was no correlation between DSQ and the Bazaz scores, and nor between the MDADI and the Bazaz scores. Time between T1 and T2 ranged from 8 to 37 days. Three patients conducted T2 in less time than the median time, which was 14 days. There was a significant correlation between the DSQ at T1 and T2 with a correlation coefficient of r = 0.61, P < The Cronbach alpha coefficient was calculated to 0.82, indicating a very good intraclass correlation and reliability. The Bland- Altman diagram shows that a good agreement between DSQ at T1 and T2 with no systematic trend of lower values differs more than higher values or vice versa 20 ( Figure 4 ). The average EQ-5D was 0.61 (SD = 0.36). The distribution of EQ-5D showed a biphasic appearance ( Figure 5 ). For this reason, nonparametric Gamma Correlation Test was chosen for these analyses. There was a weak, but significant, correlation between the DSQ and the EQ-5D ( r = 0.27, P < 0.05), whereas no correlation could be detected with the MDADI ( r = 0.18). The Bazaz score showed a paradoxical inverse correlation with the EQ-5D with higher QOL values associated with more dysphagia ( r = 0.31, P < 0.05). The Second Validation Study Preoperatively the DSQ scores were low with a mean value of 1.4 (SD = 1.9), but 17 patients (15%) had a score of 4 or higher. At 4 weeks postoperatively, the mean value was 3.2 (SD = 2.5). Only 16 patients (14%) reported a DSQ value of 0, whereas 46 patients (41%) reported a DSQ value of 4 or higher. At 3 months, the average DSQ value had decreased to 1. 7 (SD = 2.0). Both the values at 4 weeks and at 3 months were significantly higher than the preoperative value ( P 0.05). At 1 year after surgery, the DSQ levels were back to baseline values with a mean score of 1.2 (SD = 1.7); however, 13 patients (12%) scored 4 or more, which indicates a persistent swallowing problem ( Figure 6 ). Of the 17 patients who scored 4 or more (mean = 5.5) preoperatively, 14 had improved at 1 year and scored less than 4 points (mean = 1.6), one remained the same at 4 points and 2 had deteriorated, one from 6 to 7 points and one from 4 to 5 points. There was no correlation seen between DSQ level and smoking habits at any time of observation. DISCUSSION The anterior approach is common in surgery aimed at treating degenerative disease of the cervical spine. The procedure is considered to be safe and effective, but there are a number of known complications, of which dysphagia is very frequent. 21 The incidence of dysphagia after ACSS varies with different techniques and extent of surgery, and to the bulk of the implant used. 26 Objective measurements of the ability to swallow are often not enough to identify these patients because dysphagia often is a subjective sensation of disturbance or discomfort when swallowing. Bazaz et al 10 found an incidence of 50.3% at 1 month after surgery and 12.5% at 1 year after surgery. For this study, they used a grading system in which patients were evaluated via a telephone interview and classified on a 4-grade scale. However, the Bazaz score can be criticized from several different aspects. First, it is not self-administrated; that is, it represents an interpretation of the patient s dysphagia symptoms by the therapist, which may introduce a bias. Second, as we realized after having used the score clinically, it may have too few categories and, thus, does not result in enough discrepancy between patients. Yet another problem is that difficulties swallowing solids always score worse than liquids, while patients often actually experience quite the opposite. 27, 28 Also, as far as we know, the Bazaz scoring scale has never been formally validated despite its wide use. The MDADI is validated but may have some drawbacks in this group of patients. It is a rather Figure 1. The first validation study: The distribution of the Dysphagia Short Questionnaire (DSQ) at T May 2012

4 Figure 2. The first validation study: The distribution of the M.D. Anderson Dysphagia Inventory (MDADI) at T1. extensive tool with many questions that make it cumbersome to use, and it is designed and validated for patients with dysphagia caused by a malignancy. These patients presumably have more severe dysphagia than patients after ACSS. They also represent a subgroup in which psychological factors may have an influence on their health-related QOL, whereas in ACSS patients the treatment aims at full cure and the prognosis for the dysphagia is good. Dysphagia after ACSS has been reported by several authors but with a big variation in incidence. In a previous article from Riley et al 29 with an attempt to make a meta-analysis of the literature, they concluded that better outcome tools were needed. The DSQ is a self-administered instrument, which is easy to use and interpret. The design process of our evaluation instrument ensured the extraction of relevant items and assigning these relevant weights, that is, expert opinion validation. The instrument also resulted in a fairly normal distribution in a group of patients with established dysphagia of different magnitude. The good correlation with the gold standard MDADI can be seen as a support for progressivity: low values are generated in patients with mild forms of dysphagia and higher values in patients with more severe symptoms. In the test-retest analysis, we had a significant, but not perfect, correlation. There may be several explanations to this. There may, of course, be a problem with the precision of our instrument: the repeatability is poor. However, other plausible explanations may also be considered. The first test was performed together with the SLPs at the clinic where the patient went through the investigation and tests for swallowing. After this first session, the patients may have become Figure 3. The first validation study: There was a significant correlation between the Dysphagia Short Questionnaire (DSQ) and the M.D. Anderson Dysphagia Inventory (MDADI); r = 0.59, P < Spine 999

5 Figure 4. The first validation study: The Bland-Altman diagram of the Dysphagia Short Questionnaire (DSQ) at T1 and T2, respectively, indicating a good agreement between the 2 occasions. There is no systematic error related to the magnitude of the score. more aware of their problem and therefore more focused on it at the time of retest. This hypothesis is supported by the fact that the retest values are higher than the test values. Another plausible explanation is a temporal fluctuation of the dysphagia symptoms. Even though the patients were supposed to have chronic stationary problems, the severity may not be exactly the same 2 weeks later. However, we only included patients with dysphagia in the first validation study. If we had performed it instead on a group of patients after ACSS, we would have encountered several problems. First of all, the dysphagia would not have been stationary, which would have influenced the test-retest investigation as the patients could be expected to improve their symptoms during the 2-week test period, and depending on time after surgery, a varying group of patients would be entirely without dysphagia symptoms, scoring 0 on both tests, and thus improving the correlation coefficient considerably. Too many patients with zero scores would also have contributed to considerable floor effects. The fact that we saw a correlation within a group of individuals with various degree of dysphagia supports the validity and strength of the score. We could see a weak correlation to the QOL outcome instrument EQ-5D in the first validation study. However, the study population has chronic problems, many of which are caused by malignant diseases, and they may have many factors apart from the dysphagia that may negatively affect the QOL. This was also evident with the MDADI, which, despite being constructed with the aim to measure the impact of dysphagia on QOL and emotional status, failed to correlate with EQ-5D. On the contrary, the Bazaz score is probably Figure 5. The first validation study: The biphasic distribution of the EQ-5D indicates 2 groups of patients with different affection of their health-related quality of life May 2012

6 Figure 6. The second validation study: Box-andwhisker diagram of DSQ values. Center line equals the median. Box and whiskers show quartiles and range, respectively. not very reliable. Apart from the criticism mentioned earlier, it not only fails to correlate with both MDADI and the DSQ, but also shows a paradoxically inverted correlation with the EQ-5D, which would indicate an improved QOL with worse dysphagia. In the second part of the study, we found that at least 15% of the patients experienced a not negligible dysphagia before surgery, but it is unclear whether this is the expected prevalence of dysphagia in a normal population or whether it is a consequence of the degenerative disc disease. However, the fact that the majority of these patients improved after surgery would support the latter interpretation. Also, other authors have reported the possible coupling of dysphagia and cervical spondylosis; for example, sometimes patients with cervical spondylosis may develop large ventral osteophytes, which may cause dysphagia, 30, 31 which was not the case in our study population as this would have rendered exclusion. On average, the DSQ values did not significantly differ between preoperative levels and at 1 year after surgery, which erroneously could be interpreted that surgery does not cause chronic dysphagia. Some individuals with preoperative dysphagia improve, while others deteriorate and remained with symptoms at 1 year. In the short term, it seems clear that ACSS causes dysphagia in almost all patients, but the symptoms gradually resolve during the first few months, and at 1 year only a small fraction remains. The symptoms, on a group level, were considerably milder after ACSS compared to the patients in the first validation group who also had more stationary problems. CONCLUSION As the DSQ correlates with the MD Anderson Dysphagia Inventory, a gold standard for dysphagia measurement, we can conclude that it measures dysphagia quantitatively. Its reproducibility is also satisfactory. We can, thus, consider the DSQ a validated tool for measuring dysphagia. Postoperative dysphagia is very common after ACSS, particularly during the first weeks after surgery. The level of postoperative dysphagia is generally mild, and it resolves over time. At 1 year only a small fraction of patients remain with chronic problem. Whether additional improvement occurs with more time cannot be answered in the present study. Key Points Dysphagia is a sensation of difficulties with swallowing and cannot always be assessed with objective methods of measurement. For assessment of postoperative dysphagia after ACSS, there is a need for an easy-to-use validated tool. The DSQ fulfills these criteria. Postoperative dysphagia after ACSS is common but seems to be transient in most patients. Dysphagia may be a symptom of cervical spondylosis. Acknowledgments The authors thank Eva Gulle for excellent help with administration and handling of data. They also thank Jonathan Mair and Lars-Olov Skeppholm for help with translations and Johan Bring for statistical advice. Supplemental digital content is available for this article. Direct URL citations appearing in the printed text are provided in the HTML and PDF version of this article on the journal s web site ( References 1. Martin RE, Neary MA, Diamant NE. Dysphagia following anterior cervical spine surgery. Dysphagia 1997 ; 12 : 2 8 ; discussion Papavero L, Heese O, Klotz-Regener V, et al. The impact of esophagus retraction on early dysphagia after anterior cervical surgery: does a correlation exist? Spine 2007 ; 32 : Spine

7 3. Fogel GR, McDonnell MF. Surgical treatment of dysphagia after anterior cervical interbody fusion. Spine J 2005 ; 5 : Caspar W, Barbier DD, Klara PM. Anterior cervical fusion and Caspar plate stabilization for cervical trauma. Neurosurgery 1989 ; 25 : Apfelbaum RI, Kriskovich MD, Haller JR. On the incidence, cause, and prevention of recurrent laryngeal nerve palsies during anterior cervical spine surgery. Spine (Phila Pa 1976) 2000 ; 25 : Schneeberger AG, Boos N, Schwarzenbach O, et al. Anterior cervical interbody fusion with plate fixation for chronic spondylotic radiculopathy: a 2- to 8-year follow-up. J Spinal Disord 1999 ; 12 : ; discussion Stewart M, Johnston RA, Stewart I, et al. Swallowing performance following anterior cervical spine surgery. Br J Neurosurg 1995 ; 9 : Smith-Hammond CA, New KC, Pietrobon R, et al. Prospective analysis of incidence and risk factors of dysphagia in spine surgery patients: comparison of anterior cervical, posterior cervical, and lumbar procedures. Spine 2004 ; 29 : Frempong-Boadu A, Houten JK, Osborn B, et al. Swallowing and speech dysfunction in patients undergoing anterior cervical discectomy and fusion: a prospective, objective preoperative and postoperative assessment. J Spinal Disord Tech 2002 ; 15 : Bazaz R, Lee MJ, Yoo JU. Incidence of dysphagia after anterior cervical spine surgery: a prospective study. Spine 2002 ; 27 : Tervonen H, Niemela M, Lauri ER, et al. Dysphonia and dysphagia after anterior cervical decompression. J Neurosurg Spine 2007 ; 7 : Winslow CP, Winslow TJ, Wax MK. Dysphonia and dysphagia following the anterior approach to the cervical spine. Arch Otolaryngol Head Neck Surg 2001 ; 127 : Yue WM, Brodner W, Highland TR. Persistent swallowing and voice problems after anterior cervical discectomy and fusion with allograft and plating: a 5- to 11-year follow-up study. Eur Spine J 2005 ; 14 : Lee SK, Lee GY, Wong GT. Prolonged and severe dysphagia following anterior cervical surgery. J Clin Neurosci 2004 ; 11 : Chen AY, Frankowski R, Bishop-Leone J, et al. The development and validation of a dysphagia-specific quality-of-life questionnaire for patients with head and neck cancer: the M. D. Anderson dysphagia inventory. Arch Otolaryngol Head Neck Surg 2001 ; 127 : McHorney CA, Robins J, Lomax K, et al. The SWAL-QOL and SWAL-CARE outcomes tool for oropharyngeal dysphagia in adults: III. Documentation of reliability and validity. Dysphagia 2002 ; 17 : Rabin R, de Charro F. EQ-5D: a measure of health status from the EuroQol Group. Ann Med 2001 ; 33 : Burstrom K, Johannesson M, Diderichsen F. Swedish population health-related quality of life results using the EQ-5D. Qual Life Res 2001 ; 10 : Brooks R. EuroQol: the current state of play. Health Policy 1996 ; 37 : Dewitte K, Fierens C, Stocki D, et al. Application of the Bland- Altman plot for interpretation of method-comparison studies: a critical investigation of its practice. Clin Chem 2002 ; 48 : ; author reply Daniels AH, Riew KD, Yoo JU, et al. Adverse events associated with anterior cervical spine surgery. J Am Acad Orthop Surg 2008 ; 16 : Lee MJ, Bazaz R, Furey CG, et al. Risk factors for dysphagia after anterior cervical spine surgery: a two-year prospective cohort study. Spine J 2007 ; 7 : Heese O, Schroder F, Westphal M, et al. Intraoperative measurement of pharynx/esophagus retraction during anterior cervical surgery. Part I: pressure. Eur Spine J 2006 ; 15 : Mendoza-Lattes S, Clifford K, Bartelt R, et al. Dysphagia following anterior cervical arthrodesis is associated with continuous, strong retraction of the esophagus. J Bone Joint Surg Am 2008 ; 90 : McAfee PC, Cappuccino A, Cunnigham BW, et al. Lower incidence of dysphagia with cervical arthroplasty compared with ACDF in a prospective randomized clinical trial. J Spinal Disord Tech 23 : Chin KR, Eiszner JR, Adams SB Jr. Role of plate thickness as a cause of dysphagia after anterior cervical fusion. Spine (Phila Pa 1976) 2007 ; 32 : Clave P, et al. The effect of bolus viscosity on swallowing function in neurogenic dysphagia. Aliment Pharmacol Ther 2006 ; 24 : Saitoh E, Shibata S, Matsuo K, et al. Chewing and food consistency: effects on bolus transport and swallow initiation. Dysphagia 2007 ; 22 : Riley LH III, Vaccaro AR, Dettori JR, et al. Postoperative dysphagia in anterior cervical spine surgery. Spine (Phila Pa 1976) 2010 ; 35 : S Kibel SM, Johnson PM. Surgery for osteophyte-induced dysphagia. J Laryngol Otol 1987 ; 101 : Oppenlander ME, et al. Dysphagia due to anterior cervical hyperosteophytosis. Surg Neurol 2009 ; 72 : ; discussion May 2012