Factors influencing postoperative symptom improvement in hiatal hernia patients and construction of a clinical prediction model

doi:10.21203/rs.3.rs-3635175/v1

Download PDF

Research Article

Factors influencing postoperative symptom improvement in hiatal hernia patients and construction of a clinical prediction model

https://doi.org/10.21203/rs.3.rs-3635175/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Purpose: Our purpose was to analyze the factors influencing the improvement of symptoms postoperatively in patients with a hiatal hernia (HH), and build a clinical prediction model.

Methods: The records of 121 patients with a HH who received surgery at the Sixth Affiliated Hospital of Sun Yat-Sen University from April 2019 to October 2022 were retrospectively reviewed. Patients were divided into a good improvement group (88 patients) and a poor improvement group (33 patients) according to the improvement of symptoms postoperatively. Patient demographic and clinical data, and high-resolution manometry (HRM) data were extracted from the records, and data of the 2 groups were compared by univariate analysis. Statistically significant factors (p<0.05) were incorporated into a multivariate logistic regression model to determine independent factors of postoperative symptom improvement. A clinical prediction model and a nomogram were constructed. Receiver operating characteristic (ROC) curve and calibration curve analysis were used to predict the accuracy of the model. Decision curve analysis (DCA) was used to evaluate the clinical applicability of the model.

Results: Multivariate logistic regression analysis showed that non-smoking history and level of distal contraction integral (DCI) were independent factors affecting postoperative symptom improvement of HH patients. The accuracy of the test scoring system shows that the C index is 0.878.

Conclusion:The Improvement of symptoms postoperatively in HH patients is affected by smoking history and DCI level. The clinical prediction model established based on these results has good efficacy in predicting the outcomes of patients with a HH who undergo surgery.

Hiatal hernia

Laparoscopes

Nomogram

Clinical prediction model

A hiatal hernia (HH) is when the contents of the abdominal cavity (most commonly the stomach, omentum, and colon) enter the chest through the esophageal hiatus, as a result of destruction of the normal anatomical structure at the junction of the esophagus and stomach [1]. Symptoms can be divided into gastroesophageal reflux symptoms, which include a burning sensation, acid heartburn, nausea and retching, substernal fullness or pain, and vague upper abdominal pain, and extra-esophageal symptoms such as reflux cough, reflux laryngitis, reflux asthma, and reflux dental erosion [2]. In recent years, the incidence of HH has been increasing, primarily as a result of increased awareness, and the incidence is about 4% in China and 5–15% in other countries [3]. Medical treatments are generally not effective, and can lead to a long-term psychological and economic burden for the patients [4].

Surgery is the primary treatment for a HH, and laparoscopic HH repair has the advantages of effective repair of the diaphragm defect and a quick recovery. In addition, then combined with fundoplication the anti-reflux effect is enhanced [5]. The degree of improvement of symptoms postoperatively is an important index of the effectiveness of surgery. Thus, the purpose of this study was to determine factors that influence the improvement of symptoms postoperatively in patients with a HH, and establish a predictive model to determine which patients will benefit from surgery, and which patients are not likely to benefit from surgery.

2.1 Patients

The records of a total of 145 patients with a HH who underwent surgery at the Sixth Affiliated Hospital of Sun Yat-Sen University from April 2019 to October 2022 were retrospectively reviewed. Inclusion criteria for the study were: (1) Gastroscopy, computed tomography (CT), and high-resolution manometry (HRM) completed before surgery, and the main diagnosis was HH; (2) The patient underwent laparoscopic HH repair and fundoplication under tracheal intubation and general anesthesia. Patients lost to follow-up and those with incomplete data of interest were excluded. Finally, 121 patients, 83 males and 38 females, were included in the analysis. Based on symptoms postoperatively, the patients were divided into a good symptom improvement group (88 patients) and a poor symptom improvement group (33 patients).

2.2 Methods

All patients underwent routine examinations, including an electrocardiogram and blood tests, before surgery. All patients also received HRM and 24-hour gastric acid monitoring to determine esophageal motility and acid reflux, and gastroscopy to determine esophageal and gastric fundus conditions. Preoperative hypertension and diabetes were controlled prior to surgery. All patients underwent laparoscopic HH repair and fundoplication under tracheal intubation and general anesthesia. Postoperative symptom improvement was divided into 3 categories: 1) significant improvement (complete remission of symptoms); 2) Partial improvement (partial remission of symptoms); and 3) no improvement (no remission of symptoms). A postoperative telephone follow-up interview was conducted in December 2022. Patients who answered "significant improvement" were placed in the good symptom improvement group, and those who answered "partial improvement" or "no improvement" were placed in the poor symptom improvement group.

Demographic and clinical data of the 2 groups were compared by univariate analysis. Data included in the univariate analysis were age, sex, body mass index (BMI), smoking history (daily smoking > 1 cigarette, and cumulative time > 6 months was defined as smoking history), drinking history (consumption of alcohol more than 1 time per week was defined as a history of alcohol consumption), surgical history, and the presence or absence of esophagitis, gastritis, and Barrett esophagus. Other data included were DM score and the proportion of total acid reflux identified in 24-hour gastric acid monitoring, and relevant parameters in HRM, Lastly, mesh type used in the operation, and fundus folding angle were included in the univariate analysis.

2.3 Statistical analysis

Continuous data were expressed as mean ± standard deviation, and compared by t-test. Categorical data were expressed as count and percentage, and compared by χ² test. Statistically significant variables identified in the univariate analysis were incorporated into a multivariate logistic regression model, and factors significantly related to postoperative symptom improvement were identified. A clinical prediction model and nomogram were then developed. The area under the receiver operating characteristic (ROC) curve (AUC) was used to evaluate the predictive value of the model. A calibration curve was used to evaluate the calibration degree of the model, and decision curve analysis (DCA) was used to evaluate the clinical applicability of the model. Statistical analyses were performed with SPSS version 26.0 and R Studio version 4.2.3 software. A value of p < 0.05 was considered to indicate a statistically significant difference.

3.1 Comparison of general data between the 2 groups

Univariate analysis of variables in the good symptom improvement group and the poor symptom improvement identified 9 variables with statistically significant differences between groups. The variables were sex, smoking history, esophageal resting pressure (ERP), esophageal resting pressure-lower esophageal sphincter average (ERP_LSE), esophageal motility_distal amplitude EM_da), distal contraction integral (DCI), LES_PE, esophageal length (EL), presence of esophagitis. There was no statistically significant differences in other indexes compared between the 2 groups (Table 1).

Table 1

Clinical characteristics of the patients
Clinical characteristics	level	good symptom improvement group(n = 88)	poor symptom improvement group(n = 33)	P
Gender (%)	Female	18 (20.5)	20 (60.6)	<0.001
	Male	70 (79.5)	13 (39.4)
Age (mean (SD))		52.91 (15.38)	58.97 (14.09)	0.051
Smoking history (%)	No	56 (63.6)	29 (87.9)	0.009
	Yes	32 (36.4)	4 (12.1)
Drinking history (%)	No	63 (71.6)	28 (84.8)	0.133
	Yes	25 (28.4)	5 (15.2)
Fundus folding angle (%)	180°	42 (47.7)	20 (60.6)	0.124
	270°	23 (26.1)	3 (9.1)
	360°	23 (26.1)	10 (30.3)
The proportion of total acid reflux (mean (SD))		8.25 (9.94)	7.14 (7.14)	0.559
DM score (mean (SD))		34.45 (36.03)	25.85 (23.10)	0.205
Barrett esophagus (%)	No	76 (86.4)	31 (93.9)	0.246
	Yes	12 (13.6)	2 (6.1)
ERP (mmHg, mean (SD))		5.72 (3.16)	7.23 (4.29)	0.036
ERP_LES (mmHg, mean (SD))		13.24 (6.68)	16.98 (9.67)	0.017
ERP_UES (mmHg, mean (SD))		66.13 (42.92)	58.74 (34.32)	0.377
LES_RP (mmHg, mean (SD))		5.71 (3.39)	7.04 (4.45)	0.082
UES_RP (mmHg, mean (SD))		10.58 (16.18)	11.50 (12.89)	0.77
EM_da (mmHg, mean (SD))		51.81 (26.56)	86.91 (32.92)	<0.001
EM_pwd (sec, mean (SD))		3.16 (0.82)	3.41 (0.66)	0.127
EM_pwov (cm/sec, mean (SD))		4.71 (3.35)	3.67 (1.39)	0.085
DCI (mmHgscm, mean (SD))		721.01 (518.10)	1633.89 (775.17)	<0.001
LES_PE (mmHg, mean (SD))		42.93 (3.44)	39.55 (3.33)	<0.001
LES_iap (mmHg, mean (SD))		2.72 (1.13)	2.63 (1.11)	0.709
EL (cm, mean (SD))		23.70 (2.74)	21.92 (2.85)	0.002
Esophagitis (%)	No	39 (44.3)	23 (69.7)	0.013
	Yes	49 (55.7)	10 (30.3)
Gastritis (%)	No	8 (9.1)	2 (6.1)	0.59
	Yes	80 (90.9)	31 (93.9)
Surgical history (%)	No	86 (97.7)	31 (93.9)	0.299
	Yes	2 (2.3)	2 (6.1)
Mesh type (%)	Synthetic mesh	60 (68.2)	27 (81.8)	0.31
	Biological mesh	12 (13.6)	2 (6.1)
	Not use	16 (18.2)	4 (12.1)
BMI (kg/m², mean (SD))		24.45 (4.15)	23.55 (3.32)	0.269
ERP = Esophageal resting pressure; LES = lower esophageal sphincter; UES = upper esophageal sphincter; ERP_LES = Esophageal resting pressure(LES respiratory average); ERP_UES = Esophageal resting pressure (UES average); LES_RP = LES residual pressure; UES_RP = UES residual pressure; EM_da = Esophageal motility (distal amplitude); EM_pwd = Esophageal motility (peristaltic wave duration); EM_pwov = Esophageal motility (peristaltic wave onset velocity); DCI = Distal contraction integral; LES_PE = Proximal end of LES; LES_iap = Intra-abdominal portion of LES; EL = Esophageal length

3.2 Multivariate logistic regression analysis of factors influencing postoperative symptom improvement

The variables with differences which were statistically significant in the univariate analysis were incorporated into the multivariate logistic regression analysis model, and the results showed that there were 2 variables, smoking history and DCI level, that were significantly associated with outcomes (Table 2).

Table 2

**Multivariate Logistic regression of patients with poor postoperative symptom improvement**
	B	SE	Wald	P	OR	95%CI
Gender	0.263	0.926	0.080	0.777	1.301	0.212 ~ 7.987
Smoking history	-2.234	0.940	5.644	0.018	0.107	0.017 ~ 0.676
IRP	-0.072	0.122	0.345	0.557	0.931	0.733 ~ 1.182
ERP_LES	0.079	0.063	1.572	0.210	1.083	0.956 ~ 1.226
EM_da	-0.002	0.017	0.011	0.917	0.998	0.965 ~ 1.032
DCI	0.003	0.001	8.971	0.003	1.003	1.001 ~ 1.005
LES_PE	-0.285	0.177	2.585	0.108	0.752	0.532 ~ 1.064
EL	-0.054	0.188	0.082	0.775	0.948	0.656 ~ 1.370
Esophagitis	1.258	0.812	2.403	0.121	3.520	0.717 ~ 17.280

3.3 Construction of a clinical prediction model for poor improvement of symptoms postoperatively

Based on the results of the multivariate logistic regression analysis, a clinical prediction model of poor improvement of symptoms after surgery was constructed, and a nomogram was developed based on the model (Fig. 1). The corresponding values of each risk variable can be scored based on the nomogram, and the total score obtained by adding the scores of the variables can estimate the probability of poor improvement of symptoms postoperatively.

3.4 ROC curve and calibration curve to predict the accuracy of the model

The variables with significant differences in the multivariate analysis were further analyzed by ROC curve analysis to examine the accuracy of the predictive model (Fig. 2). The results showed that the AUC was 0.878, indicating that the model had strong discriminative ability. The Calibration curve of the model was shown to coincidence with the actual curve, which indicates that the actual value is in good agreement with the predicted value (Fig. 3).

3.5 Clinical applicability of the prediction model

The clinical applicability of the model was evaluated by DCA. The results of DCA (Fig. 4) showed that the blue line of the model was higher than the other 2 lines, indicating that the net benefit of the model was significantly higher than the 2 extreme conditions, that is, patients could benefit from the model and it has good clinical applicability.

A HH is a common, chronic disease of the digestive tract with complex etiology and high incidence. The condition is also associated with severe complications such as bleeding, ulceration and erosion caused by reflux, which may lead to esophageal stenosis, and malignant transformation leading to esophageal adenocarcinoma [6, 7]. Long-term reflux has a serious negative impact on a patient’s life, and many patients worry about the effect of surgical treatment, so they often suffer from insomnia and anxiety [8]. Laparoscopic hernia repair combined with fundoplication has become the gold standard treatment for a HH, and has the advantages of a large field of vision, minimal surgical trauma, a remarkable anti-reflux effect, and fast recovery [9]. Postoperative symptom improvement is the primary concern of patients. Of the patients included in this study, 33 patients had poor improvement of symptoms after surgery. Thus, we developed a model to predict factors associated with a poor improvement of symptoms and the result of the model can be used to determine whether patients need surgical treatment.

Nomograms are widely used to illustrate the outcomes of predictive models, and to identify a patient’s prognosis or outcome based on clinical variables [10]. Nomograms can be highly accurate, and can help clinicians to make better plans to manage individual patients [11]. The prediction model established in this study was shown to have good accuracy by ROC curve analysis and other methods.

In this study, the highest score nomogram score was for DCI; the greater the DCI, the higher the probability of poor improvement of symptoms. HRM is a common method for evaluating esophageal motility, and the anti-reflux barrier, and it is widely used in the detection and assessment of esophageal motility disorders [12, 13]. DCI is an important parameter in HRM, and refers to the pressure × duration × length of contraction in the smooth muscle of esophagus, and the measure is used to judge the intensity of esophageal contractions. The higher the DCI, the higher the intensity of esophageal contraction and the stronger anti-reflux activity [14]. Savarino et al. [15] found that the lower the contractile ability of the esophagus body, the weaker the clearance ability of the esophagus after reflux, thus making reflux symptoms more serious. Our results showed that a higher DCI was associated with a higher probability of poor improvement of symptoms. One of the probable reasons is that DCI indicates the intensity of esophageal contraction. As mentioned above, the larger the value, the greater the ability to clear the reflux. This suggests that some of the symptoms of a patient, such as acid reflux and heartburn, which may be caused by the weakening of esophageal dynamic, are not obvious before surgery. However, the operation itself does not change the intensity of esophageal contraction, which leads to poor improvement in postoperative symptoms.

Our results also showed that no smoking history was also an independent risk factor for poor improvement in symptoms. Study has shown that smoking is an independent risk factor for gastroesophageal reflux [16.] Moreover, gastroesophageal reflux symptoms are more obvious in long-term smokers than in short-term smokers. Smoking also reduces the pressure of the lower esophageal sphincter [17], and the frequency of gastroesophageal reflux increases significantly. This leads to the reflux of stomach contents to the lower esophagus, resulting in the destruction of the barrier between the esophagus and the stomach, and further causes reflux, heartburn, and other symptoms. Compared with smokers, patients without a history of smoking had less reflux and heartburn before surgery, and less overall symptoms after surgery, but the improvement in symptoms was less significant than in smokers.

Based on an AUC = 0.878, and the good alignment of the calibration curve with the actual curve, we can see that the accuracy of the model developed in this study is high. However, this study still has some limitations. The sample size is small, and the study was done at a single center. A multi-center study is needed to verify the accuracy of the results. There were no objective indicators of improvement. Improvement of symptoms was based on the subjective responses of patients during a telephone interview, which can be subject to inaccuracies.

In summary, this study established a clinical prediction model for postoperative symptom improvement in HH patients. Patients can decide whether surgery is necessary or not based on the predicted results of this model. ROC curve, calibration curve and DCA all show that this model has good predictive efficacy, and can also provide evidence for clinicians to make decisions in clinical practice.

Ethics approval and consent to participate

This study has been approved by the Sixth Affiliated Hospital of Sun Yat-sen University (approval number: E2023126). Written informed consent was obtained from the patient for participating in the study.

Consent for publication

Not applicable

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Conflict of interest

The authors have no relevant financial or non-financial interests to disclose

Funding

Supported by Nation Ker Clinical Discipline

Acknowledgement

The authors would like to thank the patients for their willingness to participate in this study.

Authors contributions

Haonan Huang and Ning Ma designed the research. Haonan Huang、Tao Ma and Chuangxiong Liu collected and analyzed the data. Enming Huang and Fuxing Tang performed the literature search. Haonan Huang and Tao Ma drafted the article. Taicheng Zhou supervised the study. All authors were involved in writing the paper. All authors contributed to the article and approved the submitted version.

Kim, Paul et al. “Hiatal hernia classification-Way past its shelf life.” Surgery vol. 170,2 (2021): 642-643.
Kahrilas PJ, Kim HC, Pandolfino JE. Approaches to the diagnosis and grading of hiatal hernia. Best Pact Res Clin Gastroenterol. 2008;22(4):601-616.
Wang Z, Bright T, Irvine T, et al. Outcome for asymptomatic recurrence following laparoscopic repair of very large hiatus hernia. J Gastrointest Surg. 2015;19:1385–1390.
Zaman JA, Lidor AO. The optimal approach to symptomatic para-esophageal hernia repair: important technical considerations. Curr Gastroenterol Rep. 2016;18:53.
Francesco T, Daniele M, Alexandros I, et al. A new mechanism of gastroesophageal reflux in hiatal hernia documented by high-resolution impedance manometry: a case report[J]. Ann Gastroenterol, 2016, 29(4): 548-550.
Menon S, Trudgill N. Risk factors in the aetiology of hiatus hernia: a meta-analysis. Eur J Gastroenterol Hepatol. 2011;23:133–138.
Herb Ella F, Schlattmann F, Patti MG. Antireflux surgery and barrett esophagus: myth or reality[J]. World J Surg, 2018, 42(6): 1798-1802.
Roman S, Kahrilas PJ. The diagnosis and management of hiatus hernia. BMJ. 2014;349:g6154
Desiderio J, Trastulli S, Ricci F, et al. Robotic Nissen fundoplication for gastro-esophageal reflux disease with hiatal hernia (with video) [J]. J Visc Surg, 2016, 153(2): 145-146.
Collins GS, Reitsma JB, Altman DG, et al. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ, 2015, 350: g7594
Altman DG, Vergouwe Y, Royston P, et al. Prognosis and prognostic research: validating a prognostic model. BMJ, 2009, 338: b605
VARDAR R，KESKIN M. Indications of 24－h esophageal pH monitoring，capsule pH monitoring，combined pH monitoring with multichannel impedance，esophageal manometry，radiology and scintigraphy in gastroesophageal reflux disease? J］.Turk J Gastroenterol，2017，28( Suppl 1) : S16－S21．
SCHUTYSER W, CRUYT L, VULSTEKE J B, et al. The role of high-resolution manometry in the assessment of upper gastrointestinal involvement in systemic sclerosis: a systematic review[J]. Clin Rheumatol, 2020, 39(1):149-157.
Kahrilas PJ，Bredenoord AJ, Fox M, et al. The Chicago Classification of esophageal
Savarino E, Gemignani L, Pohl D, et al. Oesophageal motility and bolus transit abnormalities increase in parallel with the severity of gastro-oesophageal reflux disease. Aliment Pharmacol Ther. 2011 Aug;34(4):476-86.
T T，Tran T L T，et al. Clinical characteristics and risk factors of gastroesophageal reflux disease in vietnamese patients with upper gastrointestinal symptoms undergoing esophagogastroduodenoscopy[J]. JGH Open，2021，5（5）：580-584. ［54］
Gyawali, C Prakash, and Ronnie Fass. Management of Gastroesophageal Reflux Disease.” Gastroenterology vol. 154,2 (2018): 302-318.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
04 Jul, 2024
Reviews received at journal
29 Jun, 2024
Reviewers agreed at journal
20 Jun, 2024
Reviews received at journal
03 Apr, 2024
Reviewers agreed at journal
13 Mar, 2024
Reviewers invited by journal
10 Mar, 2024
Editor assigned by journal
29 Feb, 2024
Editor invited by journal
20 Nov, 2023
Submission checks completed at journal
20 Nov, 2023
First submitted to journal
19 Nov, 2023

You are reading this latest preprint version

Factors influencing postoperative symptom improvement in hiatal hernia patients and construction of a clinical prediction model

Status:

Version 1

Abstract

Figures

1. Introduction

2. Patients and Methods

2.1 Patients

2.2 Methods

2.3 Statistical analysis

3. Results

3.1 Comparison of general data between the 2 groups

3.2 Multivariate logistic regression analysis of factors influencing postoperative symptom improvement

3.3 Construction of a clinical prediction model for poor improvement of symptoms postoperatively

3.4 ROC curve and calibration curve to predict the accuracy of the model

3.5 Clinical applicability of the prediction model

4. Discussion

Declarations

References

Additional Declarations

Status:

Version 1