on 20-Apr-2022 (Wed)

Flow of Causation

The flow of association is symmetric, whereas the flow of causation is not. Under the causal edges assumption (Assumption 3.3), causation only flows in a single direction. Causation only flows along directed paths. Association flows along any path that does not contain an immorality

Descendants of Colliders

Conditioning on descendants of a collider also induces association in between the parents of the collider.

The incidence of, and mortality from, IE has not changed substantially over the past 30 years.1

Without treatment, IE is usually lethal. Even with surgical treatment (used is about 50% of patients) and antibiotics, the mean in-hospital mortality is as high as 15–20%, and 1-year mortality is up to 40%.1

Often, a diagnosis of IE is established only after a lag phase of 1 month owing to nonspecific symptoms, and patients have to be treated for a further 40 days in hospital.

The incidence of IE is 1.4–12.7 per 100,000 personyears.8–12 The global epidemiology of IE differs considerably between high-income countries (HIC) and low-income and middle-income countries (LMIC),1 with the most-striking difference being the up to twentyfold increased prevalence of IE (17–77%) resulting from rheumatic heart disease in LMIC.9,13

Furthermore, patients with IE in LMIC are often very young, typically presenting aged 20–40 years.

In LMIC, the aetiological pathogen is more (or equally) likely to be a Streptococcus species than a Staphylococcus species.9

Staphylococci, streptococci, and enterococci are responsible for infections in >80% of patients with IE,4 including Streptococcus mutans20 and Streptococcus sanguinis,21,22 which are typically associated with dental plaques and caries.

In 2012, Yew and Murdoch stated that the current state of IE in LMIC now reflects that of IE in HIC in the middle of the twentieth century.9

These temporal trends in HIC12,23–25 are exemplified by three 1-year, populationbased surveys (from 1991, 1999, and 2008) performed in three French regions and including a total of 11 million inhabitants (aged ≥20 years) and 993 expertly-validated cases of IE.26

The incidence of IE remained stable over time, with 35 (95% CI 31–39), 33 (95 CI 30–37), and 32 (95% CI 28–35) cases per million of the population in 1991, 1999, and 2008, respectively.26

The incidence of IE from oral streptococci did not increase, either in the whole patient population (8.1 [95% CI 6.4–10.1], 6.3 [95% CI 4.8–8.1], and 6.3 [95% CI 4.9–8.0]) or in patients with pre-existing native valve disease.26

The incidence of IE from S. aureus did not significantly change in the patient population as a whole (5.2 [95% CI 3.9–6.8], 6.8 [95% CI 5.3–8.6], and 8.2 [95% CI 6.6–10.2]; P = 0.228), but did significantly increase in the subgroup of patients without previously diagnosed native valve disease (1.6 [95% CI 0.9–2.7], 3.7 [95% CI 2.6–5.1], and 4.1 [95% CI 3.0–5.6]; P = 0.012).26

A worrisome trend is the increase in methicillinresistant S. aureus and the emergence of daptomycin resistance in staphylococci, enterococci, and oral streptococci,2,27 whose clinical relevance for patients with IE has been discussed previously.2

The National Institute for Health and Care Excellence in the UK went a step further (not without raising ethical and legal concerns36) and, in March 2008, recommended the complete cessation of antibiotic prophylaxis for IE, regardless of risk classification.

Over the next 2 years, this recommendation resulted in a 78.6% reduction (P <0.001) in the prescription of antibiotic prophylaxis, from a mean of 10,277 (SD 1,068) prescriptions per month to 2,292 (SD 176) prescriptions per month.37 Noninferiority testing showed no significant increase in either the incidence of IE (≤9.3%) or IE-related, in-hospital mortality (≤12.3%).37

Therefore, the concept of restrictive antibiotic prophylaxis seems to work

Pathogens deploy similar virulence strategies, including adherence to host structures, biofilm formation, host-cell invasion, aggression, and intracellular persistence.

Upon adherence, staphylococci can proliferate and form a biofilm, which is a bacterial aggregate embedded within a matrix consisting mainly of self-produced and host-produced polysaccharides and proteins.

The formation of a biofilm provides many advantages for the bacteria, including tight adherence, nutrition, and protection against the immune system and antimicrobial treatments.38

Adherence to host cells, such as endothelial cells, is also a prerequisite for host-cell invasion. Some bacterial pathogens, including S. aureus, have been increasingly recognised as facultative intracellular pathogens.39,40 From an intracellular location, the bacteria can induce multiple proinflammatory and cytotoxic effects that cause tissue destruction and enable the bacteria to enter structures deep inside the tissue. Furthermore, the intracellular location provides shelter from the host immune system and from antimicrobial treatments—which might explain chronic and therapy-refractory infections.

The steps from valve adherence to endocarditis include valve colonization and maturation of the vegetation (Box 1 and Figure 1).44 The pathogenic link is spontaneous deposition of platelet–fibrin aggregates on abnormal valves and on injured (Figure 2a) or inflamed (Figure 2b) cardiac endothelium, thereby forming a ‘nonbacterial thrombotic endocarditis’ (NBTE), an ideal seeding ground for circulating micro-organisms.

NBTE formation is thought to require cardiac endothelial injury or an inflamed endothelium, combined with a hypercoagulable state. NBTEs are found in 1.3% of individuals at post-mortem examination, characteristically at the valvular coaptation line on the atrial surface of the mitral or tricuspid valves, or on the ventricular surface of the aortic or pulmonary valves, the sites where vegetations are found in patients with IE.

Increasing age, malignancy, disseminated intravascular coagulation, uraemia, valvular heart disease, intracellular catheters, and systemic lupus erythematosus (among other factors) predispose individuals to NBTE.41

During bacteraemia, blood flow through a narrowed valvular orifice deposits bacteria maximally at the low-pressure sink immediately beyond an orifice, or at the site where a jet stream strikes a surface. Thrombotic material is deposited at the same site, which brings bacteria and NBTE together.41

This mechanical view of bacteria–NBTE interaction on injured valvular endothelium is complemented by biochemical processes.4 Any excoriation of the endothelium results in direct contact between the blood and subendothelial host components, including proteins of the extracellular matrix, thromboplastin, and tissue factor, which trigger blood coagulation.

The coagulum that forms on damaged endothelium contains large quantities of fibrinogen–fibrin, fibronectin, plasma proteins, and platelet proteins. Pathogens associated with IE avidly bind to these structures and colonize them during transient bacteraemia.42 In turn, adherent bacteria attract and activate blood monocytes to produce increased levels of tissue factor and cytokines.47

Conclusions 62 The time interval between IE first symptoms and diagnosis is closely related to the IE clinical 63 presentation, patient characteristics and causative microorganism. Better prognosis reported in 64 late-diagnosed IE may be related to a higher rate of valvular surgery

Infective endocarditis, which time interval between first symptoms and diagnosis was 72 less than one month, were mainly due to Staphylococcus aureus in France

Staphylococcus aureus infective endocarditis were associated with septic shock, 75 transient ischemic attack or stroke and higher mortality rates than infective 76 endocarditis due to other bacteria or infective endocarditis, which time interval 77 between first symptoms and diagnosis was more than one month.

Infective endocarditis, which time interval between first symptoms and diagnosis was 80 more than one month, were accounting for one quarter of all infective endocarditis in 81 our study and were associated with vertebral osteomyelitis and an higher rate of 82 cardiac surgery performed for hemodynamic indication than other infective 83 endocarditis.

It also has a high morbidity and cost 88 burden: its treatment requires prolonged hospitalization; one out of two patients undergoes 89 valve surgery during the acute phase of the disease; and quality of life and return to work are 90 compromised in some patients (3-4).

For decades, IE had been classified according to its mode of presentation, which led to 93 consider acute, subacute and chronic IE (5); without treatment, IE is a uniformly fatal disease 94 and the old categories of acute, subacute and chronic disease only referred to the time it was 95 anticipated to take before the patient would die

classification of IE is now multifaceted, taking into account predisposing factors (native 100 valve, prosthetic valve, intracavitary devices), the source of acquisition (community-acquired, 101 healthcare-related), as well as the patient's background (intravenous drug user, elderly), with 102 some overlap between the different classifications (1)

However, taking into account the time interval between the first symptom and the date 106 of diagnosis of IE may hold an interest in terms of diagnostic and prognostic assessment of 107 individual patients

Furthermore, revisiting the description of initial symptoms may help practitioners diagnose IE 112 earlier in the era of these newer imaging techniques.

In brief, this survey had been conducted in seven regions 125 of France (Paris, Lorraine, Rhône-Alpes, Franche-Comte, Marne, Ille-et-Vilaine, Languedoc-126 Roussillon), a population pool of 16 million inhabitants, during a 12-month-period. During 127 this period, all IE cases that were diagnosed in adult patients, before or after their referral to 128 hospital, were reported. A standardized case report form (CRF) was prospectively filled out 129 during the study and each reported case was then validated by an adjudication committee. All 130 IE that were not classified as definite according to modified-Duke criteria (9) were excluded 131 from further analysis

The data of 486 patients with definite IE were analyzed; of these, 356 patients had 202 community-acquired, 105 had nosocomial, and 14 non-nosocomial healthcare-related IE; the 203 presumed mode of acquisition was unknown in 11 patients

Time interval between initial symptoms and diagnosis 206 Most patients (362 representing 74.5% of the entire cohort) had an early-diagnosed IE while 207 124 (25.5%) had a late-diagnosed IE. Among the 235 early-diagnosed IE patients with 208 available calendar date of the first symptom onset, the time interval between diagnosis of IE 209 and first symptoms was less than 7 days in 70.2% of the patients, between 7 and 14 days in 210 17.5% and above 14 days in 12.3%. Of note, in 42 of the 124 late-diagnosed IE patient group 211 (33.9%), first IE symptoms occurred more than 3 months before diagnosis

The clinical characteristics of IE according to the time to diagnosis are described in Table I.

Among the 130 Staphylococcus aureus IE, 119 221 (91.5%) occurred in early-diagnosed IE patients group.

Among the 46 coagulase negative 222 staphylococcus IE, 34 (73.9%) occurred in early-diagnosed IE patients group (Figure 1) 223 (supplementary Table I)

Enterococci and group D streptococci were less frequently observed 224 in early-diagnosed IE patients group, whereas pyogenic streptococci, S. pneumoniae, and S. agalactiae were almost exclusively observed in early-diagnosed IE. Other Streptococcus 226 species were equally distributed between early-diagnosed IE and late-diagnosed IE patients 227 groups (supplementary Table I)

Other Streptococcus 226 species were equally distributed between early-diagnosed IE and late-diagnosed IE patients 227 groups (supplementary Table I)

Factors independently associated with early-diagnosed IE were female sex (OR = 1.8; 229 95% CI [1.0-3.0]), the presence of a prosthetic valve (OR= 2.6; 95% CI [1.4-5.0]) and 230 staphylococci as the causative pathogen (OR=3.7; 95% CI [2.2-6.2])

Symptoms occurring before or at the time of IE diagnosis are reported in Table II. Fever, 234 severe sepsis/ septic shock, and nausea were more frequently observed in early-diagnosed IE 235 than in late-diagnosed IE

Mean C Reactive Protein was higher in early-diagnosed IE than in 236 late-diagnosed IE. Weight loss and fatigue were less frequently observed in early-diagnosed 237 IE than in late-diagnosed IE.

Rates and types of theoretical indications for surgery were different according to groups. A 241 theoretical indication for valve surgery was less frequent in early-diagnosed IE patients. 242 Valve surgery was also less frequently performed (whether for heart failure or embolism 243 prevention) in early-diagnosed IE than in late-diagnosed IE (Table III).

Among early-diagnosed IE group, Staphylococcus aureus species mainly accounted for in-255 hospital or one-year mortalities, as well as for the presence of septic shock or transient 256 ischemic attack or stroke (Table III). Results were the same when the 11 late-diagnosed IE 257 due to Staphylococcus aureus (supplementary Table I) were removed from analysis (data not 258 shown).

In hospital mortality was higher in early-diagnosed IE patients than in late-diagnosed 253 IE patients (25.1 vs 16.1%) such was one-year mortality (51.9% vs 17.7%) (Table III)

Diagnosis of IE was established over one month after the beginning of symptoms 265 in 25% of patients, and as long as 3 months in 8 %

The first subgroup is composed of patients for whom 278 the infectious and/or inflammatory manifestations of IE are prominent. These IE are mainly 279 due to virulent microorganisms such as Staphylococcus aureus and pyogenic streptococci. 280 Infectious manifestations (fever, septic shock...) lead to a sound presentation, and early 281 diagnosis and care. As reported by others, this acute presentation is associated with a poor 282 prognosis, with mortality rates over twice as high as in late-diagnosed IE (14). The 283 overrepresentation of Staphylococcus aureus in this sub-group explains the high proportion of 284 nosocomial infections, and of prosthetic valve IE.

This high proportion of patients with prosthetic valve in the early-diagnosed IE (which is an independent associated factor for 286 early-diagnosed IE) is also probably due to an earlier evocation of the possibility of IE in case 287 of symptom occurrence in these patients clearly recognized at high incidence of IE

IE prognosis 290 in this subgroup of patients seems to be more related to control of the bacterial infection than 291 to valve dysfunction. These data support the interest of an empiric antibiotic strategy active 292 against Staphylococcus aureus and “virulent streptococci” in patients with acute IE, pending 293 for blood culture results

The second sub-group of patients included in the early-diagnosed EI 294 group has a clinical and microbiological profile which is quite similar to that of late-295 diagnosed IE patients. It probably represents IE which has been diagnosed rapidly after the 296 onset of first symptoms despite a less symptomatic presentation, due to more specific initial 297 symptoms and/or greater practitioner attentiveness

The late-diagnosed IE group accounted for one-quarter of all definite IE, and were 299 frequently associated with weight loss, and asthenia; late-diagnosed IE were mainly due to 300 non-virulent microorganisms such as oral or digestive streptococci on native valve diseases. 301 These data first suggested that intravenous ampicillin could be the drug of choice for 302 empirical treatment of late-diagnosed IE in the context of this study. Moreover, these data 303 also suggested that health education of patients with native valve disease could reduce the 304 time interval between symptoms like asthenia and diagnosis or between dental procedure and 305 diagnosis. Interestingly, fever was absent in more than 25% of cases (Table II); clinicians 306 should keep in mind the diagnosis of subacute IE and look for heart murmur abnormalities 307 when faced with asthenia or weight loss in patients with or without previous IE predisposing 308 cardiac conditions even without fever (15).

In fact, the diagnosis of subacute IE still remains 309 difficult due to this non-specific and polymorphic clinical presentation. This is illustrated by the long time interval before diagnosis (more than 3 months after the beginning of symptoms) 311 in some patients. No clinical sign reported here was specific enough to help the clinician 312 easily make the diagnosis of IE. This long time interval before diagnosis of IE is associated 313 with a high rate of valve destruction, which had time to occur, and with a high rate of 314 indications for hemodynamic surgery, which was finally performed in most of the patients. 315 This assertion is confirmed by the data of DeSimone and colleagues, which provides evidence 316 of a higher diagnosis delay and a higher surgery rate in euthermic endocarditis than in febrile 317 endocarditis (15)

First, the determination of initial 332 symptoms has been obviously made a posteriori by patients and practitioners (but 333 prospectively in the study) and could be affected by recall bias

Furthermore, as most of these 334 symptoms are non-specific, it is difficult to ascertain that they were really related to IE

Second, we did not take into account any microorganism virulence factors, which could differ 336 within microorganism species according to strain and could be responsible for the diversity of 337 IE presentations

In the present report, the time interval to diagnosis of IE is closely related to the types 339 of IE clinical presentation, themselves closely related to patient characteristics, 340 microorganism virulence and capacity to induce severe inflammatory response syndrome, and 341 practitioner propensity for considering the possibility of IE diagnosis

Prospective cohort study of 2781 adults with definite IE who were admitted to 58 hospitals in 25 coun- tries from June 1, 2000, through September 1, 2005

Althoughfirst described in the mid-16th century, the Gulstonian lectures by Osler 1-3 to the Royal College of Physicians in 1885 created the impetus for the systematic study of IE.

The current in-hospital mortal- ity rate for patients with IE is 15% to 20%, 5,16 with1-yearmortality approaching 40%. 16,18,19

Sites had to meet the following criteria: (1) minimum enrollment of 12 cases per year in a center with access to cardiac sur- gery; (2) patient identification procedures in place to ensure consecutive enrollment and to minimize ascertainment bias 21 ; (3) high-quality data, including query resolution; and (4) institutional review board and/or ethics committee approval or waiver based on local standards

A total of 3284 patients were enrolled into ICE-PCS, of whom 2781 had definite IE by the modified Duke criteria ( Table 1). 22

A case report form of 275 variables was developed by the ICE group according to standard definitions. 21,23,24

Most of the patients in the cohort (72.1%) had native valve IE, and most patients (77.0%) were admitted within 1 month of the initial signs of ill- ness.

The most common underlying condition was dia- betes mellitus (16.2%), but 9.9% of the South American patients had diabetes, compared with 26.7% of North American patients. Similarly, less than 5% of patients from outside North America were receiving hemodialysis, com- pared with 20.8% of North American patients.

De- generative valve disease (eg, significant mitral [43.3%] and/or aortic [26.3%] valve regurgitation) was the most frequent native valve predisposing factor.

Valvu- lar predisposing conditions also included the presence of a prosthetic valve in 618 patients (22.2%)

Chronic intravenous access was as common as intra- venous drug use in the overall cohort; 148 of 244 pa- tients (60.7%) in this study with chronic intravenous access were from North America (Table 2)

linical and laboratory findings on admission are pre- sented in Table 3. The classic signs that are often con- sidered diagnostic for IE were infrequent.

In 2756 of the 2781 patients (99.1%), blood samples were cultured to determine the causative microorganism.

Of the 310 patients (11.1%) with negative blood culture yields, 192 (61.9%) had received antibiotics within 7 days of the blood culture.

Of the 2781 patients, 277 (10.0%) had cul- tures and serologic tests that were negative for IE

Gram-positive organisms were predominant (81.5%), with Staphylococcus aureus accounting for 31.2% of all infections

Positive serologic tests for Coxiella burnetii were reported for 27 patients (17 from Europe, 2 from North America, 1 from South America, and 7 from other regions), but only 9 were re- ported to have reciprocal antibody titers of more than 800.

Similarly, 22 patients had positive serologic tests for Bar- tonella species (18 from Europe, 1 from South America, and 3 from other regions), but only 3 were reported to have reciprocal antibody titers of more than 800. One case of infection was due to Tropheryma whippelii.

Staphylococcus aureus was the most common organ- ism in 3 of 4 regions, whereas viridans group strepto- cocci were the most common organisms isolated from patients in South America

The microbial causes of IE varied with location of acqui- sition, with a higher proportion who had staphylococ- cal IE and a lower proportion who had viridans strepto- coccal IE among those with health care–associated IE.

Among patients with community-acquired infection, 34.3% had staphylococcal IE and 22.7% had viridans streptococcal IE, whereas the corresponding figures for nosocomial infection were 69.8% and 0.8%, respec- tively, and for nonnosocomial health care–associated in- fection were 67.4% and 4.3%, respectively

Echocardiography was used in most patients (99.2%). More than one-half (59.0%) of the patients had undergone transthoracic andtransesophageal echocardiography. Of the 2781patients, 87.1%had echocardiographic evidenceof veg- etation, whereasnew, significant valvularregurgitation was discovered in 63.8% of patients

Congestive heart failure was the most common compli- cation in all regions ( Table 6). In general, the highest com- plication rates occurred in North America and Europe

Abscesswas the most common paravalvular complication(14.4% ofpatients), whereas 34.1% of patients with prosthetic valve IE had evidence of a prosthetic valve complication such as dehiscence or new paravalvular regurgitation

Surgical treatment was common for the en- tire cohort (48.2%), and in-hospital mortality was 17.7%

The following variables were independently asso- ciated with an increased risk of in-hospital death: involve- ment of a prosthetic valve, increasing age, radiographic pul- monary edema, S aureus infection, coagulase-negative staphylococcus infection, presence of mitral valve vegeta- tion,andparavalvularcomplications

Of the total cohort of patients with definite IE, 1174 (42.2%) had been transferred to a study hospital from an- other health care facility. Analysis of the data after exclud- ing these patients revealed few differences from analysis of the whole cohort (Tables 2, 4, and 6). Notable differ- ences were that transferred patients were more likely to un- dergo surgery (63.4% of transferred patients vs 37.1% of nontransferred patients [P⬍.001]) and were more likely to have congestive heart failure as a complication (39.3% vs 27.1% [P⬍.001])

In contrast, most patients in this investigation presented early and demonstrated few of the classic clini- cal findings traditionally associated with IE. For ex- ample, in the 1960s and 1970s, Osler nodes were re- corded in 11% to 23% and splenomegaly in 20% to 44% of patients with IE. 9,10,25,26 In our study, predisposing val- vular conditions were common but were primarily ow- ing to the presence of degenerative valve disease or a pros- thetic valve rather than rheumatic heart disease

Staphy- lococcus aureus is now the most common cause of IE in much of the world, confirming several recent investiga- tions 5,16,31 and the earlier findings of the ICE-PCS. 23

An emerging population at risk for IE consists of pa- tients with health care–associated infections. Overall, IE was attributed to a health care–related exposure in nearly 25% of the patients.

We also noted a substantially higher prevalence of S bovis–associated IE in Europe, that HACEK-associated IE was relatively uncommon in North America, and that most cases of Q fever and Bartonella-associated IE came from Europe.

For IE due to microorganisms that are difficult to culture, geographic differences may, at least partially, reflect variation in the threshold for performing addi- tional diagnostic tests. This may be the case for Q fever and Bartonella-associated IE, which often rely on sero- logic and/or nucleic acid amplification tests for diagno- sis. 36 However, it is also clear that there are geographic differences in the incidence of these 2 infections. 37

In addition, prosthetic valve IE and staphylococcal IE were also associated with an increased risk of in-hospital death, whereas there was a decreased risk associated with viri- dans streptococcal IE.

An elevated ESR was associated with a decreased risk of death, although the reason for this is unclear. Elevated ESR may be associated with more chronic infection, thereby signifying a more chronic clini- cal course.

Although study sites spanned all non-Antarctic con- tinents, there was a heavy weighting toward wealthy coun- tries in Europe, North America, and Australasia, with few sites in Asia and Africa

Blocked path definition

Definition 3.3 (blocked path) A path between nodes 𝑋 and 𝑌 is blocked by a (potentially empty) conditioning set 𝑍 if either of the following is true:

1. Along the path, there is a chain · · · → 𝑊 → · · · or a fork · · · ← 𝑊 → · · ·, where 𝑊 is conditioned on (𝑊 ∈ 𝑍).

2. There is a collider 𝑊 on the path that is not conditioned on ( 𝑊 ∉ 𝑍 ) and none of its descendants are conditioned on (de(𝑊) * 𝑍)

Definition 3.4 (d-separation) Two (sets of) nodes 𝑋 and 𝑌 are d-separated by a set of nodes 𝑍 if all of the paths between (any node in) 𝑋 and (any node in) 𝑌 are blocked by 𝑍

Source: Pearl (1988), Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference

If all the paths between two nodes 𝑋 and 𝑌 are blocked, then we say that 𝑋 and 𝑌 are d-separated.

if there exists at least one path between 𝑋 and 𝑌 that is unblocked, then we say that 𝑋 and 𝑌 are d-connected.